Authors: Kyunghwan Shim, Jaewoong Yun, Shinkook Choi
Abstract: Multi-head self-attention (MSA) is a key component of Vision Transformers (ViTs), which have achieved great success in various vision tasks. However, their high computational cost and memory footprint hinder their deployment on resource-constrained devices. Conventional pruning approaches can only compress and accelerate the MSA module using head pruning, although the head is not an atomic unit. To address this issue, we propose a novel graph-aware neuron-level pruning method, Structured Neuron-level Pruning (SNP). SNP prunes neurons with less informative attention scores and eliminates redundancy among heads. Specifically, it prunes graphically connected query and key layers having the least informative attention scores while preserving the overall attention scores. Value layers, which can be pruned independently, are pruned to eliminate inter-head redundancy. Our proposed method effectively compresses and accelerates Transformer-based models for both edge devices and server processors. For instance, the DeiT-Small with SNP runs 3.1$\times$ faster than the original model and achieves performance that is 21.94\% faster and 1.12\% higher than the DeiT-Tiny. Additionally, SNP combine successfully with conventional head or block pruning approaches. SNP with head pruning could compress the DeiT-Base by 80\% of the parameters and computational costs and achieve 3.85$\times$ faster inference speed on RTX3090 and 4.93$\times$ on Jetson Nano.
Authors: Nagabhushan Somraj, Kapil Choudhary, Sai Harsha Mupparaju, Rajiv Soundararajan
Abstract: Designing a 3D representation of a dynamic scene for fast optimization and rendering is a challenging task. While recent explicit representations enable fast learning and rendering of dynamic radiance fields, they require a dense set of input viewpoints. In this work, we focus on learning a fast representation for dynamic radiance fields with sparse input viewpoints. However, the optimization with sparse input is under-constrained and necessitates the use of motion priors to constrain the learning. Existing fast dynamic scene models do not explicitly model the motion, making them difficult to be constrained with motion priors. We design an explicit motion model as a factorized 4D representation that is fast and can exploit the spatio-temporal correlation of the motion field. We then introduce reliable flow priors including a combination of sparse flow priors across cameras and dense flow priors within cameras to regularize our motion model. Our model is fast, compact and achieves very good performance on popular multi-view dynamic scene datasets with sparse input viewpoints. The source code for our model can be found on our project page: https://nagabhushansn95.github.io/publications/2024/RF-DeRF.html.
URLs: https://nagabhushansn95.github.io/publications/2024/RF-DeRF.html.
Authors: Jean-Paul Ainam, Erim Yanik, Rahul Rahul, Taylor Kunkes, Lora Cavuoto, Brian Clemency, Kaori Tanaka, Matthew Hackett, Jack Norfleet, Suvranu De
Abstract: Endotracheal intubation (ETI) is an emergency procedure performed in civilian and combat casualty care settings to establish an airway. Objective and automated assessment of ETI skills is essential for the training and certification of healthcare providers. However, the current approach is based on manual feedback by an expert, which is subjective, time- and resource-intensive, and is prone to poor inter-rater reliability and halo effects. This work proposes a framework to evaluate ETI skills using single and multi-view videos. The framework consists of two stages. First, a 2D convolutional autoencoder (AE) and a pre-trained self-supervision network extract features from videos. Second, a 1D convolutional enhanced with a cross-view attention module takes the features from the AE as input and outputs predictions for skill evaluation. The ETI datasets were collected in two phases. In the first phase, ETI is performed by two subject cohorts: Experts and Novices. In the second phase, novice subjects perform ETI under time pressure, and the outcome is either Successful or Unsuccessful. A third dataset of videos from a single head-mounted camera for Experts and Novices is also analyzed. The study achieved an accuracy of 100% in identifying Expert/Novice trials in the initial phase. In the second phase, the model showed 85% accuracy in classifying Successful/Unsuccessful procedures. Using head-mounted cameras alone, the model showed a 96% accuracy on Expert and Novice classification while maintaining an accuracy of 85% on classifying successful and unsuccessful. In addition, GradCAMs are presented to explain the differences between Expert and Novice behavior and Successful and Unsuccessful trials. The approach offers a reliable and objective method for automated assessment of ETI skills.
Authors: Mir Rayat Imtiaz Hossain, Mennatullah Siam, Leonid Sigal, James J. Little
Abstract: The emergence of attention-based transformer models has led to their extensive use in various tasks, due to their superior generalization and transfer properties. Recent research has demonstrated that such models, when prompted appropriately, are excellent for few-shot inference. However, such techniques are under-explored for dense prediction tasks like semantic segmentation. In this work, we examine the effectiveness of prompting a transformer-decoder with learned visual prompts for the generalized few-shot segmentation (GFSS) task. Our goal is to achieve strong performance not only on novel categories with limited examples, but also to retain performance on base categories. We propose an approach to learn visual prompts with limited examples. These learned visual prompts are used to prompt a multiscale transformer decoder to facilitate accurate dense predictions. Additionally, we introduce a unidirectional causal attention mechanism between the novel prompts, learned with limited examples, and the base prompts, learned with abundant data. This mechanism enriches the novel prompts without deteriorating the base class performance. Overall, this form of prompting helps us achieve state-of-the-art performance for GFSS on two different benchmark datasets: COCO-$20^i$ and Pascal-$5^i$, without the need for test-time optimization (or transduction). Furthermore, test-time optimization leveraging unlabelled test data can be used to improve the prompts, which we refer to as transductive prompt tuning.
Authors: Deepti Hegde, Suhas Lohit, Kuan-Chuan Peng, Michael J. Jones, Vishal M. Patel
Abstract: Popular representation learning methods encourage feature invariance under transformations applied at the input. However, in 3D perception tasks like object localization and segmentation, outputs are naturally equivariant to some transformations, such as rotation. Using pre-training loss functions that encourage equivariance of features under certain transformations provides a strong self-supervision signal while also retaining information of geometric relationships between transformed feature representations. This can enable improved performance in downstream tasks that are equivariant to such transformations. In this paper, we propose a spatio-temporal equivariant learning framework by considering both spatial and temporal augmentations jointly. Our experiments show that the best performance arises with a pre-training approach that encourages equivariance to translation, scaling, and flip, rotation and scene flow. For spatial augmentations, we find that depending on the transformation, either a contrastive objective or an equivariance-by-classification objective yields best results. To leverage real-world object deformations and motion, we consider sequential LiDAR scene pairs and develop a novel 3D scene flow-based equivariance objective that leads to improved performance overall. We show our pre-training method for 3D object detection which outperforms existing equivariant and invariant approaches in many settings.
Authors: Oishee Bintey Hoque, Samarth Swarup, Abhijin Adiga, Sayjro Kossi Nouwakpo, Madhav Marathe
Abstract: Irrigation mapping plays a crucial role in effective water management, essential for preserving both water quality and quantity, and is key to mitigating the global issue of water scarcity. The complexity of agricultural fields, adorned with diverse irrigation practices, especially when multiple systems coexist in close quarters, poses a unique challenge. This complexity is further compounded by the nature of Landsat's remote sensing data, where each pixel is rich with densely packed information, complicating the task of accurate irrigation mapping. In this study, we introduce an innovative approach that employs a progressive training method, which strategically increases patch sizes throughout the training process, utilizing datasets from Landsat 5 and 7, labeled with the WRLU dataset for precise labeling. This initial focus allows the model to capture detailed features, progressively shifting to broader, more general features as the patch size enlarges. Remarkably, our method enhances the performance of existing state-of-the-art models by approximately 20%. Furthermore, our analysis delves into the significance of incorporating various spectral bands into the model, assessing their impact on performance. The findings reveal that additional bands are instrumental in enabling the model to discern finer details more effectively. This work sets a new standard for leveraging remote sensing imagery in irrigation mapping.
Authors: Deepti Hegde, Suhas Lohit, Kuan-Chuan Peng, Michael J. Jones, Vishal M. Patel
Abstract: LiDAR datasets for autonomous driving exhibit biases in properties such as point cloud density, range, and object dimensions. As a result, object detection networks trained and evaluated in different environments often experience performance degradation. Domain adaptation approaches assume access to unannotated samples from the test distribution to address this problem. However, in the real world, the exact conditions of deployment and access to samples representative of the test dataset may be unavailable while training. We argue that the more realistic and challenging formulation is to require robustness in performance to unseen target domains. We propose to address this problem in a two-pronged manner. First, we leverage paired LiDAR-image data present in most autonomous driving datasets to perform multimodal object detection. We suggest that working with multimodal features by leveraging both images and LiDAR point clouds for scene understanding tasks results in object detectors more robust to unseen domain shifts. Second, we train a 3D object detector to learn multimodal object features across different distributions and promote feature invariance across these source domains to improve generalizability to unseen target domains. To this end, we propose CLIX$^\text{3D}$, a multimodal fusion and supervised contrastive learning framework for 3D object detection that performs alignment of object features from same-class samples of different domains while pushing the features from different classes apart. We show that CLIX$^\text{3D}$ yields state-of-the-art domain generalization performance under multiple dataset shifts.
Authors: Zuowen Wang, Chang Gao, Zongwei Wu, Marcos V. Conde, Radu Timofte, Shih-Chii Liu, Qinyu Chen, Zheng-jun Zha, Wei Zhai, Han Han, Bohao Liao, Yuliang Wu, Zengyu Wan, Zhong Wang, Yang Cao, Ganchao Tan, Jinze Chen, Yan Ru Pei, Sasskia Br\"uers, S\'ebastien Crouzet, Douglas McLelland, Oliver Coenen, Baoheng Zhang, Yizhao Gao, Jingyuan Li, Hayden Kwok-Hay So, Philippe Bich, Chiara Boretti, Luciano Prono, Mircea Lic\u{a}, David Dinucu-Jianu, C\u{a}t\u{a}lin Gr\^iu, Xiaopeng Lin, Hongwei Ren, Bojun Cheng, Xinan Zhang, Valentin Vial, Anthony Yezzi, James Tsai
Abstract: This survey reviews the AIS 2024 Event-Based Eye Tracking (EET) Challenge. The task of the challenge focuses on processing eye movement recorded with event cameras and predicting the pupil center of the eye. The challenge emphasizes efficient eye tracking with event cameras to achieve good task accuracy and efficiency trade-off. During the challenge period, 38 participants registered for the Kaggle competition, and 8 teams submitted a challenge factsheet. The novel and diverse methods from the submitted factsheets are reviewed and analyzed in this survey to advance future event-based eye tracking research.
Authors: Rui Deng, Tianpei Gu
Abstract: Reconstructing degraded images is a critical task in image processing. Although CNN and Transformer-based models are prevalent in this field, they exhibit inherent limitations, such as inadequate long-range dependency modeling and high computational costs. To overcome these issues, we introduce the Channel-Aware U-Shaped Mamba (CU-Mamba) model, which incorporates a dual State Space Model (SSM) framework into the U-Net architecture. CU-Mamba employs a Spatial SSM module for global context encoding and a Channel SSM component to preserve channel correlation features, both in linear computational complexity relative to the feature map size. Extensive experimental results validate CU-Mamba's superiority over existing state-of-the-art methods, underscoring the importance of integrating both spatial and channel contexts in image restoration.
Authors: Yiqun Xie, Zhihao Wang, Weiye Chen, Zhili Li, Xiaowei Jia, Yanhua Li, Ruichen Wang, Kangyang Chai, Ruohan Li, Sergii Skakun
Abstract: Foundation models, i.e., very large deep learning models, have demonstrated impressive performances in various language and vision tasks that are otherwise difficult to reach using smaller-size models. The major success of GPT-type of language models is particularly exciting and raises expectations on the potential of foundation models in other domains including satellite remote sensing. In this context, great efforts have been made to build foundation models to test their capabilities in broader applications, and examples include Prithvi by NASA-IBM, Segment-Anything-Model, ViT, etc. This leads to an important question: Are foundation models always a suitable choice for different remote sensing tasks, and when or when not? This work aims to enhance the understanding of the status and suitability of foundation models for pixel-level classification using multispectral imagery at moderate resolution, through comparisons with traditional machine learning (ML) and regular-size deep learning models. Interestingly, the results reveal that in many scenarios traditional ML models still have similar or better performance compared to foundation models, especially for tasks where texture is less useful for classification. On the other hand, deep learning models did show more promising results for tasks where labels partially depend on texture (e.g., burn scar), while the difference in performance between foundation models and deep learning models is not obvious. The results conform with our analysis: The suitability of foundation models depend on the alignment between the self-supervised learning tasks and the real downstream tasks, and the typical masked autoencoder paradigm is not necessarily suitable for many remote sensing problems.
Authors: Dillon Lohr, Michael J. Proulx, Oleg Komogortsev
Abstract: This paper performs the crucial work of establishing a baseline for gaze-driven authentication performance to begin answering fundamental research questions using a very large dataset of gaze recordings from 9202 people with a level of eye tracking (ET) signal quality equivalent to modern consumer-facing virtual reality (VR) platforms. The size of the employed dataset is at least an order-of-magnitude larger than any other dataset from previous related work. Binocular estimates of the optical and visual axes of the eyes and a minimum duration for enrollment and verification are required for our model to achieve a false rejection rate (FRR) of below 3% at a false acceptance rate (FAR) of 1 in 50,000. In terms of identification accuracy which decreases with gallery size, we estimate that our model would fall below chance-level accuracy for gallery sizes of 148,000 or more. Our major findings indicate that gaze authentication can be as accurate as required by the FIDO standard when driven by a state-of-the-art machine learning architecture and a sufficiently large training dataset.
Authors: Thomas Monninger, Vandana Dokkadi, Md Zafar Anwar, Steffen Staab
Abstract: Autonomous driving requires an accurate representation of the environment. A strategy toward high accuracy is to fuse data from several sensors. Learned Bird's-Eye View (BEV) encoders can achieve this by mapping data from individual sensors into one joint latent space. For cost-efficient camera-only systems, this provides an effective mechanism to fuse data from multiple cameras with different views. Accuracy can further be improved by aggregating sensor information over time. This is especially important in monocular camera systems to account for the lack of explicit depth and velocity measurements. Thereby, the effectiveness of developed BEV encoders crucially depends on the operators used to aggregate temporal information and on the used latent representation spaces. We analyze BEV encoders proposed in the literature and compare their effectiveness, quantifying the effects of aggregation operators and latent representations. While most existing approaches aggregate temporal information either in image or in BEV latent space, our analyses and performance comparisons suggest that these latent representations exhibit complementary strengths. Therefore, we develop a novel temporal BEV encoder, TempBEV, which integrates aggregated temporal information from both latent spaces. We consider subsequent image frames as stereo through time and leverage methods from optical flow estimation for temporal stereo encoding. Empirical evaluation on the NuScenes dataset shows a significant improvement by TempBEV over the baseline for 3D object detection and BEV segmentation. The ablation uncovers a strong synergy of joint temporal aggregation in the image and BEV latent space. These results indicate the overall effectiveness of our approach and make a strong case for aggregating temporal information in both image and BEV latent spaces.
Authors: Qing En, Yuhong Guo
Abstract: Medical image segmentation typically demands extensive dense annotations for model training, which is both time-consuming and skill-intensive. To mitigate this burden, exemplar-based medical image segmentation methods have been introduced to achieve effective training with only one annotated image. In this paper, we introduce a novel Cross-model Mutual learning framework for Exemplar-based Medical image Segmentation (CMEMS), which leverages two models to mutually excavate implicit information from unlabeled data at multiple granularities. CMEMS can eliminate confirmation bias and enable collaborative training to learn complementary information by enforcing consistency at different granularities across models. Concretely, cross-model image perturbation based mutual learning is devised by using weakly perturbed images to generate high-confidence pseudo-labels, supervising predictions of strongly perturbed images across models. This approach enables joint pursuit of prediction consistency at the image granularity. Moreover, cross-model multi-level feature perturbation based mutual learning is designed by letting pseudo-labels supervise predictions from perturbed multi-level features with different resolutions, which can broaden the perturbation space and enhance the robustness of our framework. CMEMS is jointly trained using exemplar data, synthetic data, and unlabeled data in an end-to-end manner. Experimental results on two medical image datasets indicate that the proposed CMEMS outperforms the state-of-the-art segmentation methods with extremely limited supervision.
Authors: Pushkar Shukla, Dhruv Srikanth, Lee Cohen, Matthew Turk
Abstract: We propose a novel approach to mitigate biases in computer vision models by utilizing counterfactual generation and fine-tuning. While counterfactuals have been used to analyze and address biases in DNN models, the counterfactuals themselves are often generated from biased generative models, which can introduce additional biases or spurious correlations. To address this issue, we propose using adversarial images, that is images that deceive a deep neural network but not humans, as counterfactuals for fair model training. Our approach leverages a curriculum learning framework combined with a fine-grained adversarial loss to fine-tune the model using adversarial examples. By incorporating adversarial images into the training data, we aim to prevent biases from propagating through the pipeline. We validate our approach through both qualitative and quantitative assessments, demonstrating improved bias mitigation and accuracy compared to existing methods. Qualitatively, our results indicate that post-training, the decisions made by the model are less dependent on the sensitive attribute and our model better disentangles the relationship between sensitive attributes and classification variables.
Authors: Tianyi Liang, Jiangqi Liu, Sicheng Song, Shiqi Jiang, Yifei Huang, Changbo Wang, Chenhui Li
Abstract: Recent advancements in Text-to-image (T2I) generation have witnessed a shift from adapting text to fixed backgrounds to creating images around text. Traditional approaches are often limited to generate layouts within static images for effective text placement. Our proposed approach, TextCenGen, introduces a dynamic adaptation of the blank region for text-friendly image generation, emphasizing text-centric design and visual harmony generation. Our method employs force-directed attention guidance in T2I models to generate images that strategically reserve whitespace for pre-defined text areas, even for text or icons at the golden ratio. Observing how cross-attention maps affect object placement, we detect and repel conflicting objects using a force-directed graph approach, combined with a Spatial Excluding Cross-Attention Constraint for smooth attention in whitespace areas. As a novel task in graphic design, experiments indicate that TextCenGen outperforms existing methods with more harmonious compositions. Furthermore, our method significantly enhances T2I model outcomes on our specially collected prompt datasets, catering to varied text positions. These results demonstrate the efficacy of TextCenGen in creating more harmonious and integrated text-image compositions.
Authors: Dongheon Lee, Seokju Yun, Youngmin Ro
Abstract: Recently, in the super-resolution (SR) domain, transformers have outperformed CNNs with fewer FLOPs and fewer parameters since they can deal with long-range dependency and adaptively adjust weights based on instance. In this paper, we demonstrate that CNNs, although less focused on in the current SR domain, surpass Transformers in direct efficiency measures. By incorporating the advantages of Transformers into CNNs, we aim to achieve both computational efficiency and enhanced performance. However, using a large kernel in the SR domain, which mainly processes large images, incurs a large computational overhead. To overcome this, we propose novel approaches to employing the large kernel, which can reduce latency by 86\% compared to the naive large kernel, and leverage an Element-wise Attention module to imitate instance-dependent weights. As a result, we introduce Partial Large Kernel CNNs for Efficient Super-Resolution (PLKSR), which achieves state-of-the-art performance on four datasets at a scale of $\times$4, with reductions of 68.1\% in latency and 80.2\% in maximum GPU memory occupancy compared to SRFormer-light.
Authors: Xiaoyu Qiu, Hao Feng, Yuechen Wang, Wengang Zhou, Houqiang Li
Abstract: Pre-trained vision-language models (VLMs) have shown remarkable generalization capabilities via prompting, which leverages VLMs as knowledge bases to extract information beneficial for downstream tasks. However, existing methods primarily employ uni-modal prompting, which only engages a uni-modal branch, failing to simultaneously adjust vision-language (V-L) features. Additionally, the one-pass forward pipeline in VLM encoding struggles to align V-L features that have a huge gap. Confronting these challenges, we propose a novel method, Progressive Multi-modal conditional Prompt Tuning (ProMPT). ProMPT exploits a recurrent structure, optimizing and aligning V-L features by iteratively utilizing image and current encoding information. It comprises an initialization and a multi-modal iterative evolution (MIE) module. Initialization is responsible for encoding image and text using a VLM, followed by a feature filter that selects text features similar to image. MIE then facilitates multi-modal prompting through class-conditional vision prompting, instance-conditional text prompting, and feature filtering. In each MIE iteration, vision prompts are obtained from the filtered text features via a vision generator, promoting image features to focus more on target object during vision prompting. The encoded image features are fed into a text generator to produce text prompts that are more robust to class shift. Thus, V-L features are progressively aligned, enabling advance from coarse to exact classifications. Extensive experiments are conducted in three settings to evaluate the efficacy of ProMPT. The results indicate that ProMPT outperforms existing methods on average across all settings, demonstrating its superior generalization.
Authors: Suyuan Huang, Haoxin Zhang, Yan Gao, Yao Hu, Zengchang Qin
Abstract: Multimodal Large Language Models (MLLMs) have demonstrated profound capabilities in understanding multimodal information, covering from Image LLMs to the more complex Video LLMs. Numerous studies have illustrated their exceptional cross-modal comprehension. Recently, integrating video foundation models with large language models to build a comprehensive video understanding system has been proposed to overcome the limitations of specific pre-defined vision tasks. However, the current advancements in Video LLMs tend to overlook the foundational contributions of Image LLMs, often opting for more complicated structures and a wide variety of multimodal data for pre-training. This approach significantly increases the costs associated with these methods.In response to these challenges, this work introduces an efficient method that strategically leverages the priors of Image LLMs, facilitating a resource-efficient transition from Image to Video LLMs. We propose RED-VILLM, a Resource-Efficient Development pipeline for Video LLMs from Image LLMs, which utilizes a temporal adaptation plug-and-play structure within the image fusion module of Image LLMs. This adaptation extends their understanding capabilities to include temporal information, enabling the development of Video LLMs that not only surpass baseline performances but also do so with minimal instructional data and training resources. Our approach highlights the potential for a more cost-effective and scalable advancement in multimodal models, effectively building upon the foundational work of Image LLMs.
Authors: Azad Singh, Vandan Gorade, Deepak Mishra
Abstract: Self-supervised learning (SSL) has emerged as a promising technique for medical image analysis due to its ability to learn without annotations. However, despite the promising potential, conventional SSL methods encounter limitations, including challenges in achieving semantic alignment and capturing subtle details. This leads to suboptimal representations, which fail to accurately capture the underlying anatomical structures and pathological details. In response to these constraints, we introduce a novel SSL framework OPTiML, employing optimal transport (OT), to capture the dense semantic invariance and fine-grained details, thereby enhancing the overall effectiveness of SSL in medical image representation learning. The core idea is to integrate OT with a cross-viewpoint semantics infusion module (CV-SIM), which effectively captures complex, fine-grained details inherent in medical images across different viewpoints. In addition to the CV-SIM module, OPTiML imposes the variance and covariance regularizations within OT framework to force the model focus on clinically relevant information while discarding less informative features. Through these, the proposed framework demonstrates its capacity to learn semantically rich representations that can be applied to various medical imaging tasks. To validate its effectiveness, we conduct experimental studies on three publicly available datasets from chest X-ray modality. Our empirical results reveal OPTiML's superiority over state-of-the-art methods across all evaluated tasks.
Authors: Hanjing Zhou, Mingze Yin, JinTai Chen, Danny Chen, Jian Wu
Abstract: One-shot semantic segmentation aims to segment query images given only ONE annotated support image of the same class. This task is challenging because target objects in the support and query images can be largely different in appearance and pose (i.e., intra-class variation). Prior works suggested that incorporating more annotated support images in few-shot settings boosts performances but increases costs due to additional manual labeling. In this paper, we propose a novel approach for ONE-shot semantic segmentation, called Group-On, which packs multiple query images in batches for the benefit of mutual knowledge support within the same category. Specifically, after coarse segmentation masks of the batch of queries are predicted, query-mask pairs act as pseudo support data to enhance mask predictions mutually, under the guidance of a simple Group-On Voting module. Comprehensive experiments on three standard benchmarks show that, in the ONE-shot setting, our Group-On approach significantly outperforms previous works by considerable margins. For example, on the COCO-20i dataset, we increase mIoU scores by 8.21% and 7.46% on ASNet and HSNet baselines, respectively. With only one support image, Group-On can be even competitive with the counterparts using 5 annotated support images.
Authors: Haoyue Liu, Shihan Peng, Lin Zhu, Yi Chang, Hanyu Zhou, Luxin Yan
Abstract: We focus on a very challenging task: imaging at nighttime dynamic scenes. Most previous methods rely on the low-light enhancement of a conventional RGB camera. However, they would inevitably face a dilemma between the long exposure time of nighttime and the motion blur of dynamic scenes. Event cameras react to dynamic changes with higher temporal resolution (microsecond) and higher dynamic range (120dB), offering an alternative solution. In this work, we present a novel nighttime dynamic imaging method with an event camera. Specifically, we discover that the event at nighttime exhibits temporal trailing characteristics and spatial non-stationary distribution. Consequently, we propose a nighttime event reconstruction network (NER-Net) which mainly includes a learnable event timestamps calibration module (LETC) to align the temporal trailing events and a non-uniform illumination aware module (NIAM) to stabilize the spatiotemporal distribution of events. Moreover, we construct a paired real low-light event dataset (RLED) through a co-axial imaging system, including 64,200 spatially and temporally aligned image GTs and low-light events. Extensive experiments demonstrate that the proposed method outperforms state-of-the-art methods in terms of visual quality and generalization ability on real-world nighttime datasets. The project are available at: https://github.com/Liu-haoyue/NER-Net.
Authors: Wei Wu, Qingnan Fan, Shuai Qin, Hong Gu, Ruoyu Zhao, Antoni B. Chan
Abstract: Precise image editing with text-to-image models has attracted increasing interest due to their remarkable generative capabilities and user-friendly nature. However, such attempts face the pivotal challenge of misalignment between the intended precise editing target regions and the broader area impacted by the guidance in practice. Despite excellent methods leveraging attention mechanisms that have been developed to refine the editing guidance, these approaches necessitate modifications through complex network architecture and are limited to specific editing tasks. In this work, we re-examine the diffusion process and misalignment problem from a frequency perspective, revealing that, due to the power law of natural images and the decaying noise schedule, the denoising network primarily recovers low-frequency image components during the earlier timesteps and thus brings excessive low-frequency signals for editing. Leveraging this insight, we introduce a novel fine-tuning free approach that employs progressive $\textbf{Fre}$qu$\textbf{e}$ncy truncation to refine the guidance of $\textbf{Diff}$usion models for universal editing tasks ($\textbf{FreeDiff}$). Our method achieves comparable results with state-of-the-art methods across a variety of editing tasks and on a diverse set of images, highlighting its potential as a versatile tool in image editing applications.
Authors: Jingfeng Guo, Xiaohan Zhang, Baozhu Zhao, Qi Liu
Abstract: Existing neural radiance fields (NeRF)-based novel view synthesis methods for large-scale outdoor scenes are mainly built on a single altitude. Moreover, they often require a priori camera shooting height and scene scope, leading to inefficient and impractical applications when camera altitude changes. In this work, we propose an end-to-end framework, termed AG-NeRF, and seek to reduce the training cost of building good reconstructions by synthesizing free-viewpoint images based on varying altitudes of scenes. Specifically, to tackle the detail variation problem from low altitude (drone-level) to high altitude (satellite-level), a source image selection method and an attention-based feature fusion approach are developed to extract and fuse the most relevant features of target view from multi-height images for high-fidelity rendering. Extensive experiments demonstrate that AG-NeRF achieves SOTA performance on 56 Leonard and Transamerica benchmarks and only requires a half hour of training time to reach the competitive PSNR as compared to the latest BungeeNeRF.
Authors: Xunsong Li, Pengzhan Sun, Yangcen Liu, Lixin Duan, Wen Li
Abstract: The interactions between human and objects are important for recognizing object-centric actions. Existing methods usually adopt a two-stage pipeline, where object proposals are first detected using a pretrained detector, and then are fed to an action recognition model for extracting video features and learning the object relations for action recognition. However, since the action prior is unknown in the object detection stage, important objects could be easily overlooked, leading to inferior action recognition performance. In this paper, we propose an end-to-end object-centric action recognition framework that simultaneously performs Detection And Interaction Reasoning in one stage. Particularly, after extracting video features with a base network, we create three modules for concurrent object detection and interaction reasoning. First, a Patch-based Object Decoder generates proposals from video patch tokens. Then, an Interactive Object Refining and Aggregation identifies important objects for action recognition, adjusts proposal scores based on position and appearance, and aggregates object-level info into a global video representation. Lastly, an Object Relation Modeling module encodes object relations. These three modules together with the video feature extractor can be trained jointly in an end-to-end fashion, thus avoiding the heavy reliance on an off-the-shelf object detector, and reducing the multi-stage training burden. We conduct experiments on two datasets, Something-Else and Ikea-Assembly, to evaluate the performance of our proposed approach on conventional, compositional, and few-shot action recognition tasks. Through in-depth experimental analysis, we show the crucial role of interactive objects in learning for action recognition, and we can outperform state-of-the-art methods on both datasets.
Authors: Nakul Sharma, Aditay Tripathi, Anirban Chakraborty, Anand Mishra
Abstract: In this work, we study the task of sketch-guided image inpainting. Unlike the well-explored natural language-guided image inpainting, which excels in capturing semantic details, the relatively less-studied sketch-guided inpainting offers greater user control in specifying the object's shape and pose to be inpainted. As one of the early solutions to this task, we introduce a novel partial discrete diffusion process (PDDP). The forward pass of the PDDP corrupts the masked regions of the image and the backward pass reconstructs these masked regions conditioned on hand-drawn sketches using our proposed sketch-guided bi-directional transformer. The proposed novel transformer module accepts two inputs -- the image containing the masked region to be inpainted and the query sketch to model the reverse diffusion process. This strategy effectively addresses the domain gap between sketches and natural images, thereby, enhancing the quality of inpainting results. In the absence of a large-scale dataset specific to this task, we synthesize a dataset from the MS-COCO to train and extensively evaluate our proposed framework against various competent approaches in the literature. The qualitative and quantitative results and user studies establish that the proposed method inpaints realistic objects that fit the context in terms of the visual appearance of the provided sketch. To aid further research, we have made our code publicly available at https://github.com/vl2g/Sketch-Inpainting .
Authors: Cheng Shi, Sibei Yang
Abstract: Foundation models, pre-trained on a large amount of data have demonstrated impressive zero-shot capabilities in various downstream tasks. However, in object detection and instance segmentation, two fundamental computer vision tasks heavily reliant on extensive human annotations, foundation models such as SAM and DINO struggle to achieve satisfactory performance. In this study, we reveal that the devil is in the object boundary, \textit{i.e.}, these foundation models fail to discern boundaries between individual objects. For the first time, we probe that CLIP, which has never accessed any instance-level annotations, can provide a highly beneficial and strong instance-level boundary prior in the clustering results of its particular intermediate layer. Following this surprising observation, we propose $\textbf{Zip}$ which $\textbf{Z}$ips up CL$\textbf{ip}$ and SAM in a novel classification-first-then-discovery pipeline, enabling annotation-free, complex-scene-capable, open-vocabulary object detection and instance segmentation. Our Zip significantly boosts SAM's mask AP on COCO dataset by 12.5% and establishes state-of-the-art performance in various settings, including training-free, self-training, and label-efficient finetuning. Furthermore, annotation-free Zip even achieves comparable performance to the best-performing open-vocabulary object detecters using base annotations. Code is released at https://github.com/ChengShiest/Zip-Your-CLIP
Authors: Song Wang, Jiawei Yu, Wentong Li, Wenyu Liu, Xiaolu Liu, Junbo Chen, Jianke Zhu
Abstract: Semantic scene completion, also known as semantic occupancy prediction, can provide dense geometric and semantic information for autonomous vehicles, which attracts the increasing attention of both academia and industry. Unfortunately, existing methods usually formulate this task as a voxel-wise classification problem and treat each voxel equally in 3D space during training. As the hard voxels have not been paid enough attention, the performance in some challenging regions is limited. The 3D dense space typically contains a large number of empty voxels, which are easy to learn but require amounts of computation due to handling all the voxels uniformly for the existing models. Furthermore, the voxels in the boundary region are more challenging to differentiate than those in the interior. In this paper, we propose HASSC approach to train the semantic scene completion model with hardness-aware design. The global hardness from the network optimization process is defined for dynamical hard voxel selection. Then, the local hardness with geometric anisotropy is adopted for voxel-wise refinement. Besides, self-distillation strategy is introduced to make training process stable and consistent. Extensive experiments show that our HASSC scheme can effectively promote the accuracy of the baseline model without incurring the extra inference cost. Source code is available at: https://github.com/songw-zju/HASSC.
Authors: Wenhao Zhang, Jun Wang, Yong Luo, Lei Yu, Wei Yu, Zheng He
Abstract: Lip-reading is to utilize the visual information of the speaker's lip movements to recognize words and sentences. Existing event-based lip-reading solutions integrate different frame rate branches to learn spatio-temporal features of varying granularities. However, aggregating events into event frames inevitably leads to the loss of fine-grained temporal information within frames. To remedy this drawback, we propose a novel framework termed Multi-view Temporal Granularity aligned Aggregation (MTGA). Specifically, we first present a novel event representation method, namely time-segmented voxel graph list, where the most significant local voxels are temporally connected into a graph list. Then we design a spatio-temporal fusion module based on temporal granularity alignment, where the global spatial features extracted from event frames, together with the local relative spatial and temporal features contained in voxel graph list are effectively aligned and integrated. Finally, we design a temporal aggregation module that incorporates positional encoding, which enables the capture of local absolute spatial and global temporal information. Experiments demonstrate that our method outperforms both the event-based and video-based lip-reading counterparts. Our code will be publicly available.
Authors: Chongjie Si, Xuehui Wang, Xiaokang Yang, Wei Shen
Abstract: Weakly Incremental Learning for Semantic Segmentation (WILSS) leverages a pre-trained segmentation model to segment new classes using cost-effective and readily available image-level labels. A prevailing way to solve WILSS is the generation of seed areas for each new class, serving as a form of pixel-level supervision. However, a scenario usually arises where a pixel is concurrently predicted as an old class by the pre-trained segmentation model and a new class by the seed areas. Such a scenario becomes particularly problematic in WILSS, as the lack of pixel-level annotations on new classes makes it intractable to ascertain whether the pixel pertains to the new class or not. To surmount this issue, we propose an innovative, tendency-driven relationship of mutual exclusivity, meticulously tailored to govern the behavior of the seed areas and the predictions generated by the pre-trained segmentation model. This relationship stipulates that predictions for the new and old classes must not conflict whilst prioritizing the preservation of predictions for the old classes, which not only addresses the conflicting prediction issue but also effectively mitigates the inherent challenge of incremental learning - catastrophic forgetting. Furthermore, under the auspices of this tendency-driven mutual exclusivity relationship, we generate pseudo masks for the new classes, allowing for concurrent execution with model parameter updating via the resolution of a bi-level optimization problem. Extensive experiments substantiate the effectiveness of our framework, resulting in the establishment of new benchmarks and paving the way for further research in this field.
Authors: Nicolas Ugrinovic, Boxiao Pan, Georgios Pavlakos, Despoina Paschalidou, Bokui Shen, Jordi Sanchez-Riera, Francesc Moreno-Noguer, Leonidas Guibas
Abstract: We introduce MultiPhys, a method designed for recovering multi-person motion from monocular videos. Our focus lies in capturing coherent spatial placement between pairs of individuals across varying degrees of engagement. MultiPhys, being physically aware, exhibits robustness to jittering and occlusions, and effectively eliminates penetration issues between the two individuals. We devise a pipeline in which the motion estimated by a kinematic-based method is fed into a physics simulator in an autoregressive manner. We introduce distinct components that enable our model to harness the simulator's properties without compromising the accuracy of the kinematic estimates. This results in final motion estimates that are both kinematically coherent and physically compliant. Extensive evaluations on three challenging datasets characterized by substantial inter-person interaction show that our method significantly reduces errors associated with penetration and foot skating, while performing competitively with the state-of-the-art on motion accuracy and smoothness. Results and code can be found on our project page (http://www.iri.upc.edu/people/nugrinovic/multiphys/).
Authors: Qiyuan Dai, Sibei Yang
Abstract: Referring image segmentation (RIS) aims to precisely segment referents in images through corresponding natural language expressions, yet relying on cost-intensive mask annotations. Weakly supervised RIS thus learns from image-text pairs to pixel-level semantics, which is challenging for segmenting fine-grained masks. A natural approach to enhancing segmentation precision is to empower weakly supervised RIS with the image segmentation foundation model SAM. Nevertheless, we observe that simply integrating SAM yields limited benefits and can even lead to performance regression due to the inevitable noise issues and challenges in excessive focus on object parts. In this paper, we present an innovative framework, Point PrompTing (PPT), incorporated with the proposed multi-source curriculum learning strategy to address these challenges. Specifically, the core of PPT is a point generator that not only harnesses CLIP's text-image alignment capability and SAM's powerful mask generation ability but also generates negative point prompts to address the noisy and excessive focus issues inherently and effectively. In addition, we introduce a curriculum learning strategy with object-centric images to help PPT gradually learn from simpler yet precise semantic alignment to more complex RIS. Experiments demonstrate that our PPT significantly and consistently outperforms prior weakly supervised techniques on mIoU by 11.34%, 14.14%, and 6.97% across RefCOCO, RefCOCO+, and G-Ref, respectively.
Authors: Claudia Cuttano, Gabriele Rosi, Gabriele Trivigno, Giuseppe Averta
Abstract: Humans show an innate capability to identify tools to support specific actions. The association between objects parts and the actions they facilitate is usually named affordance. Being able to segment objects parts depending on the tasks they afford is crucial to enable intelligent robots to use objects of daily living. Traditional supervised learning methods for affordance segmentation require costly pixel-level annotations, while weakly supervised approaches, though less demanding, still rely on object-interaction examples and support a closed set of actions. These limitations hinder scalability, may introduce biases, and usually restrict models to a limited set of predefined actions. This paper proposes AffordanceCLIP, to overcome these limitations by leveraging the implicit affordance knowledge embedded within large pre-trained Vision-Language models like CLIP. We experimentally demonstrate that CLIP, although not explicitly trained for affordances detection, retains valuable information for the task. Our AffordanceCLIP achieves competitive zero-shot performance compared to methods with specialized training, while offering several advantages: i) it works with any action prompt, not just a predefined set; ii) it requires training only a small number of additional parameters compared to existing solutions and iii) eliminates the need for direct supervision on action-object pairs, opening new perspectives for functionality-based reasoning of models.
Authors: Jie Ma, Min Hu, Pinghui Wang, Wangchun Sun, Lingyun Song, Hongbin Pei, Jun Liu, Youtian Du
Abstract: Audio-Visual Question Answering (AVQA) is a complex multi-modal reasoning task, demanding intelligent systems to accurately respond to natural language queries based on audio-video input pairs. Nevertheless, prevalent AVQA approaches are prone to overlearning dataset biases, resulting in poor robustness. Furthermore, current datasets may not provide a precise diagnostic for these methods. To tackle these challenges, firstly, we propose a novel dataset, \textit{MUSIC-AVQA-R}, crafted in two steps: rephrasing questions within the test split of a public dataset (\textit{MUSIC-AVQA}) and subsequently introducing distribution shifts to split questions. The former leads to a large, diverse test space, while the latter results in a comprehensive robustness evaluation on rare, frequent, and overall questions. Secondly, we propose a robust architecture that utilizes a multifaceted cycle collaborative debiasing strategy to overcome bias learning. Experimental results show that this architecture achieves state-of-the-art performance on both datasets, especially obtaining a significant improvement of 9.68\% on the proposed dataset. Extensive ablation experiments are conducted on these two datasets to validate the effectiveness of the debiasing strategy. Additionally, we highlight the limited robustness of existing multi-modal QA methods through the evaluation on our dataset.
Authors: Jingyao Wang, Yunhan Tian, Yuxuan Yang, Xiaoxin Chen, Changwen Zheng, Wenwen Qiang
Abstract: Micro-expressions (MEs) are involuntary movements revealing people's hidden feelings, which has attracted numerous interests for its objectivity in emotion detection. However, despite its wide applications in various scenarios, micro-expression recognition (MER) remains a challenging problem in real life due to three reasons, including (i) data-level: lack of data and imbalanced classes, (ii) feature-level: subtle, rapid changing, and complex features of MEs, and (iii) decision-making-level: impact of individual differences. To address these issues, we propose a dual-branch meta-auxiliary learning method, called LightmanNet, for fast and robust micro-expression recognition. Specifically, LightmanNet learns general MER knowledge from limited data through a dual-branch bi-level optimization process: (i) In the first level, it obtains task-specific MER knowledge by learning in two branches, where the first branch is for learning MER features via primary MER tasks, while the other branch is for guiding the model obtain discriminative features via auxiliary tasks, i.e., image alignment between micro-expressions and macro-expressions since their resemblance in both spatial and temporal behavioral patterns. The two branches of learning jointly constrain the model of learning meaningful task-specific MER knowledge while avoiding learning noise or superficial connections between MEs and emotions that may damage its generalization ability. (ii) In the second level, LightmanNet further refines the learned task-specific knowledge, improving model generalization and efficiency. Extensive experiments on various benchmark datasets demonstrate the superior robustness and efficiency of LightmanNet.
Authors: Zeliang Ma, Song Yang, Zhe Cui, Zhicheng Zhao, Fei Su, Delong Liu, Jingyu Wang
Abstract: The new trend in multi-object tracking task is to track objects of interest using natural language. However, the scarcity of paired prompt-instance data hinders its progress. To address this challenge, we propose a high-quality yet low-cost data generation method base on Unreal Engine 5 and construct a brand-new benchmark dataset, named Refer-UE-City, which primarily includes scenes from intersection surveillance videos, detailing the appearance and actions of people and vehicles. Specifically, it provides 14 videos with a total of 714 expressions, and is comparable in scale to the Refer-KITTI dataset. Additionally, we propose a multi-level semantic-guided multi-object framework called MLS-Track, where the interaction between the model and text is enhanced layer by layer through the introduction of Semantic Guidance Module (SGM) and Semantic Correlation Branch (SCB). Extensive experiments on Refer-UE-City and Refer-KITTI datasets demonstrate the effectiveness of our proposed framework and it achieves state-of-the-art performance. Code and datatsets will be available.
Authors: Renrong Shao, Wei Zhang, Jianhua Yin, Jun Wang
Abstract: Data-free knowledge distillation (DFKD) is a promising approach for addressing issues related to model compression, security privacy, and transmission restrictions. Although the existing methods exploiting DFKD have achieved inspiring achievements in coarse-grained classification, in practical applications involving fine-grained classification tasks that require more detailed distinctions between similar categories, sub-optimal results are obtained. To address this issue, we propose an approach called DFKD-FGVC that extends DFKD to fine-grained visual categorization~(FGVC) tasks. Our approach utilizes an adversarial distillation framework with attention generator, mixed high-order attention distillation, and semantic feature contrast learning. Specifically, we introduce a spatial-wise attention mechanism to the generator to synthesize fine-grained images with more details of discriminative parts. We also utilize the mixed high-order attention mechanism to capture complex interactions among parts and the subtle differences among discriminative features of the fine-grained categories, paying attention to both local features and semantic context relationships. Moreover, we leverage the teacher and student models of the distillation framework to contrast high-level semantic feature maps in the hyperspace, comparing variances of different categories. We evaluate our approach on three widely-used FGVC benchmarks (Aircraft, Cars196, and CUB200) and demonstrate its superior performance.
Authors: Ziang Ren, Samuel Lensgraf, Alberto Quattrini Li
Abstract: Accurate localization is fundamental for autonomous underwater vehicles (AUVs) to carry out precise tasks, such as manipulation and construction. Vision-based solutions using fiducial marker are promising, but extremely challenging underwater because of harsh lighting condition underwater. This paper introduces a gradient-based active camera exposure control method to tackle sharp lighting variations during image acquisition, which can establish better foundation for subsequent image enhancement procedures. Considering a typical scenario for underwater operations where visual tags are used, we proposed several experiments comparing our method with other state-of-the-art exposure control method including Active Exposure Control (AEC) and Gradient-based Exposure Control (GEC). Results show a significant improvement in the accuracy of robot localization. This method is an important component that can be used in visual-based state estimation pipeline to improve the overall localization accuracy.
Authors: Charles Gaydon, Floryne Roche
Abstract: Knowledge of tree species distribution is fundamental to managing forests. New deep learning approaches promise significant accuracy gains for forest mapping, and are becoming a critical tool for mapping multiple tree species at scale. To advance the field, deep learning researchers need large benchmark datasets with high-quality annotations. To this end, we present the PureForest dataset: a large-scale, open, multimodal dataset designed for tree species classification from both Aerial Lidar Scanning (ALS) point clouds and Very High Resolution (VHR) aerial images. Most current public Lidar datasets for tree species classification have low diversity as they only span a small area of a few dozen annotated hectares at most. In contrast, PureForest has 18 tree species grouped into 13 semantic classes, and spans 339 km$^2$ across 449 distinct monospecific forests, and is to date the largest and most comprehensive Lidar dataset for the identification of tree species. By making PureForest publicly available, we hope to provide a challenging benchmark dataset to support the development of deep learning approaches for tree species identification from Lidar and/or aerial imagery. In this data paper, we describe the annotation workflow, the dataset, the recommended evaluation methodology, and establish a baseline performance from both 3D and 2D modalities.
Authors: Weikang Yu, Xiaokang Zhang, Samiran Das, Xiao Xiang Zhu, Pedram Ghamisi
Abstract: Change detection (CD) from remote sensing (RS) images using deep learning has been widely investigated in the literature. It is typically regarded as a pixel-wise labeling task that aims to classify each pixel as changed or unchanged. Although per-pixel classification networks in encoder-decoder structures have shown dominance, they still suffer from imprecise boundaries and incomplete object delineation at various scenes. For high-resolution RS images, partly or totally changed objects are more worthy of attention rather than a single pixel. Therefore, we revisit the CD task from the mask prediction and classification perspective and propose MaskCD to detect changed areas by adaptively generating categorized masks from input image pairs. Specifically, it utilizes a cross-level change representation perceiver (CLCRP) to learn multiscale change-aware representations and capture spatiotemporal relations from encoded features by exploiting deformable multihead self-attention (DeformMHSA). Subsequently, a masked-attention-based detection transformers (MA-DETR) decoder is developed to accurately locate and identify changed objects based on masked attention and self-attention mechanisms. It reconstructs the desired changed objects by decoding the pixel-wise representations into learnable mask proposals and making final predictions from these candidates. Experimental results on five benchmark datasets demonstrate the proposed approach outperforms other state-of-the-art models. Codes and pretrained models are available online (https://github.com/EricYu97/MaskCD).
Authors: Zhong Wang, Zengyu Wan, Han Han, Bohao Liao, Yuliang Wu, Wei Zhai, Yang Cao, Zheng-jun Zha
Abstract: Event-based eye tracking has shown great promise with the high temporal resolution and low redundancy provided by the event camera. However, the diversity and abruptness of eye movement patterns, including blinking, fixating, saccades, and smooth pursuit, pose significant challenges for eye localization. To achieve a stable event-based eye-tracking system, this paper proposes a bidirectional long-term sequence modeling and time-varying state selection mechanism to fully utilize contextual temporal information in response to the variability of eye movements. Specifically, the MambaPupil network is proposed, which consists of the multi-layer convolutional encoder to extract features from the event representations, a bidirectional Gated Recurrent Unit (GRU), and a Linear Time-Varying State Space Module (LTV-SSM), to selectively capture contextual correlation from the forward and backward temporal relationship. Furthermore, the Bina-rep is utilized as a compact event representation, and the tailor-made data augmentation, called as Event-Cutout, is proposed to enhance the model's robustness by applying spatial random masking to the event image. The evaluation on the ThreeET-plus benchmark shows the superior performance of the MambaPupil, which secured the 1st place in CVPR'2024 AIS Event-based Eye Tracking challenge.
Authors: Wu Ran, Peirong Ma, Zhiquan He, Hao Ren, Hong Lu
Abstract: Recent advances in image deraining have focused on training powerful models on mixed multiple datasets comprising diverse rain types and backgrounds. However, this approach tends to overlook the inherent differences among rainy images, leading to suboptimal results. To overcome this limitation, we focus on addressing various rainy images by delving into meaningful representations that encapsulate both the rain and background components. Leveraging these representations as instructive guidance, we put forth a Context-based Instance-level Modulation (CoI-M) mechanism adept at efficiently modulating CNN- or Transformer-based models. Furthermore, we devise a rain-/detail-aware contrastive learning strategy to help extract joint rain-/detail-aware representations. By integrating CoI-M with the rain-/detail-aware Contrastive learning, we develop CoIC, an innovative and potent algorithm tailored for training models on mixed datasets. Moreover, CoIC offers insight into modeling relationships of datasets, quantitatively assessing the impact of rain and details on restoration, and unveiling distinct behaviors of models given diverse inputs. Extensive experiments validate the efficacy of CoIC in boosting the deraining ability of CNN and Transformer models. CoIC also enhances the deraining prowess remarkably when real-world dataset is included.
Authors: Nikolina Kubiak, Armin Mustafa, Graeme Phillipson, Stephen Jolly, Simon Hadfield
Abstract: In this paper we present S3R-Net, the Self-Supervised Shadow Removal Network. The two-branch WGAN model achieves self-supervision relying on the unify-and-adaptphenomenon - it unifies the style of the output data and infers its characteristics from a database of unaligned shadow-free reference images. This approach stands in contrast to the large body of supervised frameworks. S3R-Net also differentiates itself from the few existing self-supervised models operating in a cycle-consistent manner, as it is a non-cyclic, unidirectional solution. The proposed framework achieves comparable numerical scores to recent selfsupervised shadow removal models while exhibiting superior qualitative performance and keeping the computational cost low.
Authors: Yuzhu Cai, Sheng Yin, Yuxi Wei, Chenxin Xu, Weibo Mao, Felix Juefei-Xu, Siheng Chen, Yanfeng Wang
Abstract: The burgeoning landscape of text-to-image models, exemplified by innovations such as Midjourney and DALLE 3, has revolutionized content creation across diverse sectors. However, these advancements bring forth critical ethical concerns, particularly with the misuse of open-source models to generate content that violates societal norms. Addressing this, we introduce Ethical-Lens, a framework designed to facilitate the value-aligned usage of text-to-image tools without necessitating internal model revision. Ethical-Lens ensures value alignment in text-to-image models across toxicity and bias dimensions by refining user commands and rectifying model outputs. Systematic evaluation metrics, combining GPT4-V, HEIM, and FairFace scores, assess alignment capability. Our experiments reveal that Ethical-Lens enhances alignment capabilities to levels comparable with or superior to commercial models like DALLE 3, ensuring user-generated content adheres to ethical standards while maintaining image quality. This study indicates the potential of Ethical-Lens to ensure the sustainable development of open-source text-to-image tools and their beneficial integration into society. Our code is available at https://github.com/yuzhu-cai/Ethical-Lens.
Authors: Raz Lapid, Almog Dubin, Moshe Sipper
Abstract: This paper presents RADAR-Robust Adversarial Detection via Adversarial Retraining-an approach designed to enhance the robustness of adversarial detectors against adaptive attacks, while maintaining classifier performance. An adaptive attack is one where the attacker is aware of the defenses and adapts their strategy accordingly. Our proposed method leverages adversarial training to reinforce the ability to detect attacks, without compromising clean accuracy. During the training phase, we integrate into the dataset adversarial examples, which were optimized to fool both the classifier and the adversarial detector, enabling the adversarial detector to learn and adapt to potential attack scenarios. Experimental evaluations on the CIFAR-10 and SVHN datasets demonstrate that our proposed algorithm significantly improves a detector's ability to accurately identify adaptive adversarial attacks -- without sacrificing clean accuracy.
Authors: Shouwei Ruan, Yinpeng Dong, Hanqing Liu, Yao Huang, Hang Su, Xingxing Wei
Abstract: Vision-Language Pre-training (VLP) models like CLIP have achieved remarkable success in computer vision and particularly demonstrated superior robustness to distribution shifts of 2D images. However, their robustness under 3D viewpoint variations is still limited, which can hinder the development for real-world applications. This paper successfully addresses this concern while keeping VLPs' original performance by breaking through two primary obstacles: 1) the scarcity of training data and 2) the suboptimal fine-tuning paradigms. To combat data scarcity, we build the Multi-View Caption (MVCap) dataset -- a comprehensive collection of over four million multi-view image-text pairs across more than 100K objects, providing more potential for VLP models to develop generalizable viewpoint-invariant representations. To address the limitations of existing paradigms in performance trade-offs and training efficiency, we design a novel fine-tuning framework named Omniview-Tuning (OVT). Specifically, OVT introduces a Cross-Viewpoint Alignment objective through a minimax-like optimization strategy, which effectively aligns representations of identical objects from diverse viewpoints without causing overfitting. Additionally, OVT fine-tunes VLP models in a parameter-efficient manner, leading to minimal computational cost. Extensive experiments on various VLP models with different architectures validate that OVT significantly improves the models' resilience to viewpoint shifts and keeps the original performance, establishing a pioneering standard for boosting the viewpoint invariance of VLP models.
Authors: Ziyu Shu, Zhixin Pan
Abstract: Deep image prior (DIP) proposed in recent research has revealed the inherent trait of convolutional neural networks (CNN) for capturing substantial low-level image statistics priors. This framework efficiently addresses the inverse problems in image processing and has induced extensive applications in various domains. However, as the whole algorithm is initialized randomly, the DIP algorithm often lacks stability. Thus, this method still has space for further improvement. In this paper, we propose the self-reinforcement deep image prior (SDIP) as an improved version of the original DIP. We observed that the changes in the DIP networks' input and output are highly correlated during each iteration. SDIP efficiently utilizes this trait in a reinforcement learning manner, where the current iteration's output is utilized by a steering algorithm to update the network input for the next iteration, guiding the algorithm toward improved results. Experimental results across multiple applications demonstrate that our proposed SDIP framework offers improvement compared to the original DIP method and other state-of-the-art methods.
Authors: George Retsinas, Niki Efthymiou, Petros Maragos
Abstract: Modern agricultural applications rely more and more on deep learning solutions. However, training well-performing deep networks requires a large amount of annotated data that may not be available and in the case of 3D annotation may not even be feasible for human annotators. In this work, we develop a deep learning approach to segment mushrooms and estimate their pose on 3D data, in the form of point clouds acquired by depth sensors. To circumvent the annotation problem, we create a synthetic dataset of mushroom scenes, where we are fully aware of 3D information, such as the pose of each mushroom. The proposed network has a fully convolutional backbone, that parses sparse 3D data, and predicts pose information that implicitly defines both instance segmentation and pose estimation task. We have validated the effectiveness of the proposed implicit-based approach for a synthetic test set, as well as provided qualitative results for a small set of real acquired point clouds with depth sensors. Code is publicly available at https://github.com/georgeretsi/mushroom-pose.
Authors: Zhen Han, Chaojie Mao, Zeyinzi Jiang, Yulin Pan, Jingfeng Zhang
Abstract: Given an original image, image editing aims to generate an image that align with the provided instruction. The challenges are to accept multimodal inputs as instructions and a scarcity of high-quality training data, including crucial triplets of source/target image pairs and multimodal (text and image) instructions. In this paper, we focus on image style editing and present StyleBooth, a method that proposes a comprehensive framework for image editing and a feasible strategy for building a high-quality style editing dataset. We integrate encoded textual instruction and image exemplar as a unified condition for diffusion model, enabling the editing of original image following multimodal instructions. Furthermore, by iterative style-destyle tuning and editing and usability filtering, the StyleBooth dataset provides content-consistent stylized/plain image pairs in various categories of styles. To show the flexibility of StyleBooth, we conduct experiments on diverse tasks, such as text-based style editing, exemplar-based style editing and compositional style editing. The results demonstrate that the quality and variety of training data significantly enhance the ability to preserve content and improve the overall quality of generated images in editing tasks. Project page can be found at https://ali-vilab.github.io/stylebooth-page/.
Authors: Insoo Kim, Jae Seok Choi, Geonseok Seo, Kinam Kwon, Jinwoo Shin, Hyong-Euk Lee
Abstract: As recent advances in mobile camera technology have enabled the capability to capture high-resolution images, such as 4K images, the demand for an efficient deblurring model handling large motion has increased. In this paper, we discover that the image residual errors, i.e., blur-sharp pixel differences, can be grouped into some categories according to their motion blur type and how complex their neighboring pixels are. Inspired by this, we decompose the deblurring (regression) task into blur pixel discretization (pixel-level blur classification) and discrete-to-continuous conversion (regression with blur class map) tasks. Specifically, we generate the discretized image residual errors by identifying the blur pixels and then transform them to a continuous form, which is computationally more efficient than naively solving the original regression problem with continuous values. Here, we found that the discretization result, i.e., blur segmentation map, remarkably exhibits visual similarity with the image residual errors. As a result, our efficient model shows comparable performance to state-of-the-art methods in realistic benchmarks, while our method is up to 10 times computationally more efficient.
Authors: Tommie Kerssies, Daan de Geus, Gijs Dubbelman
Abstract: Recent vision foundation models (VFMs) have demonstrated proficiency in various tasks but require supervised fine-tuning to perform the task of semantic segmentation effectively. Benchmarking their performance is essential for selecting current models and guiding future model developments for this task. The lack of a standardized benchmark complicates comparisons. Therefore, the primary objective of this paper is to study how VFMs should be benchmarked for semantic segmentation. To do so, various VFMs are fine-tuned under various settings, and the impact of individual settings on the performance ranking and training time is assessed. Based on the results, the recommendation is to fine-tune the ViT-B variants of VFMs with a 16x16 patch size and a linear decoder, as these settings are representative of using a larger model, more advanced decoder and smaller patch size, while reducing training time by more than 13 times. Using multiple datasets for training and evaluation is also recommended, as the performance ranking across datasets and domain shifts varies. Linear probing, a common practice for some VFMs, is not recommended, as it is not representative of end-to-end fine-tuning. The benchmarking setup recommended in this paper enables a performance analysis of VFMs for semantic segmentation. The findings of such an analysis reveal that pretraining with promptable segmentation is not beneficial, whereas masked image modeling (MIM) with abstract representations is crucial, even more important than the type of supervision used. The code for efficiently fine-tuning VFMs for semantic segmentation can be accessed through the project page at: https://tue-mps.github.io/benchmark-vfm-ss/.
Authors: Andrei Niculae, Andy Catruna, Adrian Cosma, Daniel Rosner, Emilian Radoi
Abstract: Surveillance footage represents a valuable resource and opportunities for conducting gait analysis. However, the typical low quality and high noise levels in such footage can severely impact the accuracy of pose estimation algorithms, which are foundational for reliable gait analysis. Existing literature suggests a direct correlation between the efficacy of pose estimation and the subsequent gait analysis results. A common mitigation strategy involves fine-tuning pose estimation models on noisy data to improve robustness. However, this approach may degrade the downstream model's performance on the original high-quality data, leading to a trade-off that is undesirable in practice. We propose a processing pipeline that incorporates a task-targeted artifact correction model specifically designed to pre-process and enhance surveillance footage before pose estimation. Our artifact correction model is optimized to work alongside a state-of-the-art pose estimation network, HRNet, without requiring repeated fine-tuning of the pose estimation model. Furthermore, we propose a simple and robust method for obtaining low quality videos that are annotated with poses in an automatic manner with the purpose of training the artifact correction model. We systematically evaluate the performance of our artifact correction model against a range of noisy surveillance data and demonstrate that our approach not only achieves improved pose estimation on low-quality surveillance footage, but also preserves the integrity of the pose estimation on high resolution footage. Our experiments show a clear enhancement in gait analysis performance, supporting the viability of the proposed method as a superior alternative to direct fine-tuning strategies. Our contributions pave the way for more reliable gait analysis using surveillance data in real-world applications, regardless of data quality.
Authors: Radu Chivereanu, Adrian Cosma, Andy Catruna, Razvan Rughinis, Emilian Radoi
Abstract: Large Language Models (LLMs) have demonstrated remarkable capabilities in various domains, including data augmentation and synthetic data generation. This work explores the use of LLMs to generate rich textual descriptions for motion sequences, encompassing both actions and walking patterns. We leverage the expressive power of LLMs to align motion representations with high-level linguistic cues, addressing two distinct tasks: action recognition and retrieval of walking sequences based on appearance attributes. For action recognition, we employ LLMs to generate textual descriptions of actions in the BABEL-60 dataset, facilitating the alignment of motion sequences with linguistic representations. In the domain of gait analysis, we investigate the impact of appearance attributes on walking patterns by generating textual descriptions of motion sequences from the DenseGait dataset using LLMs. These descriptions capture subtle variations in walking styles influenced by factors such as clothing choices and footwear. Our approach demonstrates the potential of LLMs in augmenting structured motion attributes and aligning multi-modal representations. The findings contribute to the advancement of comprehensive motion understanding and open up new avenues for leveraging LLMs in multi-modal alignment and data augmentation for motion analysis. We make the code publicly available at https://github.com/Radu1999/WalkAndText
Authors: Jan Niklas Kolf, Naser Damer, Fadi Boutros
Abstract: Face Image Quality Assessment (FIQA) estimates the utility of face images for automated face recognition (FR) systems. We propose in this work a novel approach to assess the quality of face images based on inspecting the required changes in the pre-trained FR model weights to minimize differences between testing samples and the distribution of the FR training dataset. To achieve that, we propose quantifying the discrepancy in Batch Normalization statistics (BNS), including mean and variance, between those recorded during FR training and those obtained by processing testing samples through the pretrained FR model. We then generate gradient magnitudes of pretrained FR weights by backpropagating the BNS through the pretrained model. The cumulative absolute sum of these gradient magnitudes serves as the FIQ for our approach. Through comprehensive experimentation, we demonstrate the effectiveness of our training-free and quality labeling-free approach, achieving competitive performance to recent state-of-theart FIQA approaches without relying on quality labeling, the need to train regression networks, specialized architectures, or designing and optimizing specific loss functions.
Authors: Viktoria Ehm, Maolin Gao, Paul Roetzer, Marvin Eisenberger, Daniel Cremers, Florian Bernard
Abstract: Finding correspondences between 3D shapes is an important and long-standing problem in computer vision, graphics and beyond. A prominent challenge are partial-to-partial shape matching settings, which occur when the shapes to match are only observed incompletely (e.g. from 3D scanning). Although partial-to-partial matching is a highly relevant setting in practice, it is rarely explored. Our work bridges the gap between existing (rather artificial) 3D full shape matching and partial-to-partial real-world settings by exploiting geometric consistency as a strong constraint. We demonstrate that it is indeed possible to solve this challenging problem in a variety of settings. For the first time, we achieve geometric consistency for partial-to-partial matching, which is realized by a novel integer non-linear program formalism building on triangle product spaces, along with a new pruning algorithm based on linear integer programming. Further, we generate a new inter-class dataset for partial-to-partial shape-matching. We show that our method outperforms current SOTA methods on both an established intra-class dataset and our novel inter-class dataset.
Authors: Jin Gao, Shubo Lin, Shaoru Wang, Yutong Kou, Zeming Li, Liang Li, Congxuan Zhang, Xiaoqin Zhang, Yizheng Wang, Weiming Hu
Abstract: Masked image modeling (MIM) pre-training for large-scale vision transformers (ViTs) in computer vision has enabled promising downstream performance on top of the learned self-supervised ViT features. In this paper, we question if the extremely simple ViTs' fine-tuning performance with a small-scale architecture can also benefit from this pre-training paradigm, which is considerably less studied yet in contrast to the well-established lightweight architecture design methodology with sophisticated components introduced. By carefully adapting various typical MIM pre-training methods to this lightweight regime and comparing them with the contrastive learning (CL) pre-training on various downstream image classification and dense prediction tasks, we systematically observe different behaviors between MIM and CL with respect to the downstream fine-tuning data scales. Furthermore, we analyze the frozen features under linear probing evaluation and also the layer representation similarities and attention maps across the obtained models, which clearly show the inferior learning of MIM pre-training on higher layers, leading to unsatisfactory fine-tuning performance on data-insufficient downstream tasks. This finding is naturally a guide to choosing appropriate distillation strategies during pre-training to solve the above deterioration problem. Extensive experiments on various vision tasks demonstrate the effectiveness of our observation-analysis-solution flow. In particular, our pre-training with distillation on pure lightweight ViTs with vanilla/hierarchical design (5.7M/6.5M) can achieve 79.4%/78.9% top-1 accuracy on ImageNet-1K. It also enables SOTA performance on the ADE20K semantic segmentation task (42.8% mIoU) and LaSOT visual tracking task (66.1% AUC) in the lightweight regime. The latter even surpasses all the current SOTA lightweight CPU-realtime trackers.
Authors: Han Fang, Xianghao Zang, Chao Ban, Zerun Feng, Lanxiang Zhou, Zhongjiang He, Yongxiang Li, Hao Sun
Abstract: Text-video retrieval aims to find the most relevant cross-modal samples for a given query. Recent methods focus on modeling the whole spatial-temporal relations. However, since video clips contain more diverse content than captions, the model aligning these asymmetric video-text pairs has a high risk of retrieving many false positive results. In this paper, we propose Probabilistic Token Aggregation (\textit{ProTA}) to handle cross-modal interaction with content asymmetry. Specifically, we propose dual partial-related aggregation to disentangle and re-aggregate token representations in both low-dimension and high-dimension spaces. We propose token-based probabilistic alignment to generate token-level probabilistic representation and maintain the feature representation diversity. In addition, an adaptive contrastive loss is proposed to learn compact cross-modal distribution space. Based on extensive experiments, \textit{ProTA} achieves significant improvements on MSR-VTT (50.9%), LSMDC (25.8%), and DiDeMo (47.2%).
Authors: Xianyu Chen, Ming Jiang, Qi Zhao
Abstract: Understanding how attention varies across individuals has significant scientific and societal impacts. However, existing visual scanpath models treat attention uniformly, neglecting individual differences. To bridge this gap, this paper focuses on individualized scanpath prediction (ISP), a new attention modeling task that aims to accurately predict how different individuals shift their attention in diverse visual tasks. It proposes an ISP method featuring three novel technical components: (1) an observer encoder to characterize and integrate an observer's unique attention traits, (2) an observer-centric feature integration approach that holistically combines visual features, task guidance, and observer-specific characteristics, and (3) an adaptive fixation prioritization mechanism that refines scanpath predictions by dynamically prioritizing semantic feature maps based on individual observers' attention traits. These novel components allow scanpath models to effectively address the attention variations across different observers. Our method is generally applicable to different datasets, model architectures, and visual tasks, offering a comprehensive tool for transforming general scanpath models into individualized ones. Comprehensive evaluations using value-based and ranking-based metrics verify the method's effectiveness and generalizability.
Authors: Andrei-Timotei Ardelean, Tim Weyrich
Abstract: Anomaly detection and localization in images is a growing field in computer vision. In this area, a seemingly understudied problem is anomaly clustering, i.e., identifying and grouping different types of anomalies in a fully unsupervised manner. In this work, we propose a novel method for clustering anomalies in largely stationary images (textures) in a blind setting. That is, the input consists of normal and anomalous images without distinction and without labels. What contributes to the difficulty of the task is that anomalous regions are often small and may present only subtle changes in appearance, which can be easily overshadowed by the genuine variance in the texture. Moreover, each anomaly type may have a complex appearance distribution. We introduce a novel scheme for solving this task using a combination of blind anomaly localization and contrastive learning. By identifying the anomalous regions with high fidelity, we can restrict our focus to those regions of interest; then, contrastive learning is employed to increase the separability of different anomaly types and reduce the intra-class variation. Our experiments show that the proposed solution yields significantly better results compared to prior work, setting a new state of the art. Project page: https://reality.tf.fau.de/pub/ardelean2024blind.html.
Authors: Matthias Schwab, Agnes Mayr, Markus Haltmeier
Abstract: The recent emergence of deep learning has led to a great deal of work on designing supervised deep semantic segmentation algorithms. As in many tasks sufficient pixel-level labels are very difficult to obtain, we propose a method which combines a Gaussian mixture model (GMM) with unsupervised deep learning techniques. In the standard GMM the pixel values with each sub-region are modelled by a Gaussian distribution. In order to identify the different regions, the parameter vector that minimizes the negative log-likelihood (NLL) function regarding the GMM has to be approximated. For this task, usually iterative optimization methods such as the expectation-maximization (EM) algorithm are used. In this paper, we propose to estimate these parameters directly from the image using a convolutional neural network (CNN). We thus change the iterative procedure in the EM algorithm replacing the expectation-step by a gradient-step with regard to the networks parameters. This means that the network is trained to minimize the NLL function of the GMM which comes with at least two advantages. As once trained, the network is able to predict label probabilities very quickly compared with time consuming iterative optimization methods. Secondly, due to the deep image prior our method is able to partially overcome one of the main disadvantages of GMM, which is not taking into account correlation between neighboring pixels, as it assumes independence between them. We demonstrate the advantages of our method in various experiments on the example of myocardial infarct segmentation on multi-sequence MRI images.
Authors: Gautham Vinod, Jiangpeng He, Zeman Shao, Fengqing Zhu
Abstract: Image-based methods to analyze food images have alleviated the user burden and biases associated with traditional methods. However, accurate portion estimation remains a major challenge due to the loss of 3D information in the 2D representation of foods captured by smartphone cameras or wearable devices. In this paper, we propose a new framework to estimate both food volume and energy from 2D images by leveraging the power of 3D food models and physical reference in the eating scene. Our method estimates the pose of the camera and the food object in the input image and recreates the eating occasion by rendering an image of a 3D model of the food with the estimated poses. We also introduce a new dataset, SimpleFood45, which contains 2D images of 45 food items and associated annotations including food volume, weight, and energy. Our method achieves an average error of 31.10 kCal (17.67%) on this dataset, outperforming existing portion estimation methods.
Authors: Mohammed Shaiqur Rahman, Ibne Farabi Shihab, Lynna Chu, Anuj Sharma
Abstract: In this study, we introduce DeepLocalization, an innovative framework devised for the real-time localization of actions tailored explicitly for monitoring driver behavior. Utilizing the power of advanced deep learning methodologies, our objective is to tackle the critical issue of distracted driving-a significant factor contributing to road accidents. Our strategy employs a dual approach: leveraging Graph-Based Change-Point Detection for pinpointing actions in time alongside a Video Large Language Model (Video-LLM) for precisely categorizing activities. Through careful prompt engineering, we customize the Video-LLM to adeptly handle driving activities' nuances, ensuring its classification efficacy even with sparse data. Engineered to be lightweight, our framework is optimized for consumer-grade GPUs, making it vastly applicable in practical scenarios. We subjected our method to rigorous testing on the SynDD2 dataset, a complex benchmark for distracted driving behaviors, where it demonstrated commendable performance-achieving 57.5% accuracy in event classification and 51% in event detection. These outcomes underscore the substantial promise of DeepLocalization in accurately identifying diverse driver behaviors and their temporal occurrences, all within the bounds of limited computational resources.
Authors: Israel A. Laurensi, Alceu de Souza Britto Jr., Jean Paul Barddal, Alessandro Lameiras Koerich
Abstract: Facial expression recognition is a pivotal component in machine learning, facilitating various applications. However, convolutional neural networks (CNNs) are often plagued by catastrophic forgetting, impeding their adaptability. The proposed method, emotion-centered generative replay (ECgr), tackles this challenge by integrating synthetic images from generative adversarial networks. Moreover, ECgr incorporates a quality assurance algorithm to ensure the fidelity of generated images. This dual approach enables CNNs to retain past knowledge while learning new tasks, enhancing their performance in emotion recognition. The experimental results on four diverse facial expression datasets demonstrate that incorporating images generated by our pseudo-rehearsal method enhances training on the targeted dataset and the source dataset while making the CNN retain previously learned knowledge.
Authors: Yona Falinie A. Gaus, Neelanjan Bhowmik, Brian K. S. Isaac-Medina, Toby P. Breckon
Abstract: The Segment Anything Model (SAM) is a deep neural network foundational model designed to perform instance segmentation which has gained significant popularity given its zero-shot segmentation ability. SAM operates by generating masks based on various input prompts such as text, bounding boxes, points, or masks, introducing a novel methodology to overcome the constraints posed by dataset-specific scarcity. While SAM is trained on an extensive dataset, comprising ~11M images, it mostly consists of natural photographic images with only very limited images from other modalities. Whilst the rapid progress in visual infrared surveillance and X-ray security screening imaging technologies, driven forward by advances in deep learning, has significantly enhanced the ability to detect, classify and segment objects with high accuracy, it is not evident if the SAM zero-shot capabilities can be transferred to such modalities. This work assesses SAM capabilities in segmenting objects of interest in the X-ray/infrared modalities. Our approach reuses the pre-trained SAM with three different prompts: bounding box, centroid and random points. We present quantitative/qualitative results to showcase the performance on selected datasets. Our results show that SAM can segment objects in the X-ray modality when given a box prompt, but its performance varies for point prompts. Specifically, SAM performs poorly in segmenting slender objects and organic materials, such as plastic bottles. We find that infrared objects are also challenging to segment with point prompts given the low-contrast nature of this modality. This study shows that while SAM demonstrates outstanding zero-shot capabilities with box prompts, its performance ranges from moderate to poor for point prompts, indicating that special consideration on the cross-modal generalisation of SAM is needed when considering use on X-ray/infrared imagery.
Authors: Niklas Penzel, Gideon Stein, Joachim Denzler
Abstract: Deep models trained on large amounts of data often incorporate implicit biases present during training time. If later such a bias is discovered during inference or deployment, it is often necessary to acquire new data and retrain the model. This behavior is especially problematic in critical areas such as autonomous driving or medical decision-making. In these scenarios, new data is often expensive and hard to come by. In this work, we present a method based on change penalization that takes a pre-trained model and adapts the weights to mitigate a previously detected bias. We achieve this by tuning a zero-initialized copy of a frozen pre-trained network. Our method needs very few, in extreme cases only a single, examples that contradict the bias to increase performance. Additionally, we propose an early stopping criterion to modify baselines and reduce overfitting. We evaluate our approach on a well-known bias in skin lesion classification and three other datasets from the domain shift literature. We find that our approach works especially well with very few images. Simple fine-tuning combined with our early stopping also leads to performance benefits for a larger number of tuning samples.
Authors: Tristan Piater, Niklas Penzel, Gideon Stein, Joachim Denzler
Abstract: A substantial body of research has focused on developing systems that assist medical professionals during labor-intensive early screening processes, many based on convolutional deep-learning architectures. Recently, multiple studies explored the application of so-called self-attention mechanisms in the vision domain. These studies often report empirical improvements over fully convolutional approaches on various datasets and tasks. To evaluate this trend for medical imaging, we extend two widely adopted convolutional architectures with different self-attention variants on two different medical datasets. With this, we aim to specifically evaluate the possible advantages of additional self-attention. We compare our models with similarly sized convolutional and attention-based baselines and evaluate performance gains statistically. Additionally, we investigate how including such layers changes the features learned by these models during the training. Following a hyperparameter search, and contrary to our expectations, we observe no significant improvement in balanced accuracy over fully convolutional models. We also find that important features, such as dermoscopic structures in skin lesion images, are still not learned by employing self-attention. Finally, analyzing local explanations, we confirm biased feature usage. We conclude that merely incorporating attention is insufficient to surpass the performance of existing fully convolutional methods.
Authors: Md Adnan Arefeen, Biplob Debnath, Md Yusuf Sarwar Uddin, Srimat Chakradhar
Abstract: Retrieval augmented generation (RAG) systems combine the strengths of language generation and information retrieval to power many real-world applications like chatbots. Use of RAG for combined understanding of multimodal data such as text, images and videos is appealing but two critical limitations exist: one-time, upfront capture of all content in large multimodal data as text descriptions entails high processing times, and not all information in the rich multimodal data is typically in the text descriptions. Since the user queries are not known apriori, developing a system for multimodal to text conversion and interactive querying of multimodal data is challenging. To address these limitations, we propose iRAG, which augments RAG with a novel incremental workflow to enable interactive querying of large corpus of multimodal data. Unlike traditional RAG, iRAG quickly indexes large repositories of multimodal data, and in the incremental workflow, it uses the index to opportunistically extract more details from select portions of the multimodal data to retrieve context relevant to an interactive user query. Such an incremental workflow avoids long multimodal to text conversion times, overcomes information loss issues by doing on-demand query-specific extraction of details in multimodal data, and ensures high quality of responses to interactive user queries that are often not known apriori. To the best of our knowledge, iRAG is the first system to augment RAG with an incremental workflow to support efficient interactive querying of large, real-world multimodal data. Experimental results on real-world long videos demonstrate 23x to 25x faster video to text ingestion, while ensuring that quality of responses to interactive user queries is comparable to responses from a traditional RAG where all video data is converted to text upfront before any querying.
Authors: Jiayi Liang, Haotian Liu, Hongteng Xu, Dixin Luo
Abstract: As a significant step for human face modeling, editing, and generation, face landmarking aims at extracting facial keypoints from images. A generalizable face landmarker is required in practice because real-world facial images, e.g., the avatars in animations and games, are often stylized in various ways. However, achieving generalizable face landmarking is challenging due to the diversity of facial styles and the scarcity of labeled stylized faces. In this study, we propose a simple but effective paradigm to learn a generalizable face landmarker based on labeled real human faces and unlabeled stylized faces. Our method learns the face landmarker as the key module of a conditional face warper. Given a pair of real and stylized facial images, the conditional face warper predicts a warping field from the real face to the stylized one, in which the face landmarker predicts the ending points of the warping field and provides us with high-quality pseudo landmarks for the corresponding stylized facial images. Applying an alternating optimization strategy, we learn the face landmarker to minimize $i)$ the discrepancy between the stylized faces and the warped real ones and $ii)$ the prediction errors of both real and pseudo landmarks. Experiments on various datasets show that our method outperforms existing state-of-the-art domain adaptation methods in face landmarking tasks, leading to a face landmarker with better generalizability. Code is available at https://plustwo0.github.io/project-face-landmarker}{https://plustwo0.github.io/project-face-landmarker.
URLs: https://plustwo0.github.io/project-face-landmarker, https://plustwo0.github.io/project-face-landmarker.
Authors: Christoph Reich, Oliver Hahn, Daniel Cremers, Stefan Roth, Biplob Debnath
Abstract: Resource-constrained hardware, such as edge devices or cell phones, often rely on cloud servers to provide the required computational resources for inference in deep vision models. However, transferring image and video data from an edge or mobile device to a cloud server requires coding to deal with network constraints. The use of standardized codecs, such as JPEG or H.264, is prevalent and required to ensure interoperability. This paper aims to examine the implications of employing standardized codecs within deep vision pipelines. We find that using JPEG and H.264 coding significantly deteriorates the accuracy across a broad range of vision tasks and models. For instance, strong compression rates reduce semantic segmentation accuracy by more than 80% in mIoU. In contrast to previous findings, our analysis extends beyond image and action classification to localization and dense prediction tasks, thus providing a more comprehensive perspective.
Authors: Nupur Kumari, Grace Su, Richard Zhang, Taesung Park, Eli Shechtman, Jun-Yan Zhu
Abstract: Model customization introduces new concepts to existing text-to-image models, enabling the generation of the new concept in novel contexts. However, such methods lack accurate camera view control w.r.t the object, and users must resort to prompt engineering (e.g., adding "top-view") to achieve coarse view control. In this work, we introduce a new task -- enabling explicit control of camera viewpoint for model customization. This allows us to modify object properties amongst various background scenes via text prompts, all while incorporating the target camera pose as additional control. This new task presents significant challenges in merging a 3D representation from the multi-view images of the new concept with a general, 2D text-to-image model. To bridge this gap, we propose to condition the 2D diffusion process on rendered, view-dependent features of the new object. During training, we jointly adapt the 2D diffusion modules and 3D feature predictions to reconstruct the object's appearance and geometry while reducing overfitting to the input multi-view images. Our method outperforms existing image editing and model personalization baselines in preserving the custom object's identity while following the input text prompt and the object's camera pose.
Authors: Ronghuan Wu, Wanchao Su, Kede Ma, Jing Liao
Abstract: Clipart, a pre-made graphic art form, offers a convenient and efficient way of illustrating visual content. Traditional workflows to convert static clipart images into motion sequences are laborious and time-consuming, involving numerous intricate steps like rigging, key animation and in-betweening. Recent advancements in text-to-video generation hold great potential in resolving this problem. Nevertheless, direct application of text-to-video generation models often struggles to retain the visual identity of clipart images or generate cartoon-style motions, resulting in unsatisfactory animation outcomes. In this paper, we introduce AniClipart, a system that transforms static clipart images into high-quality motion sequences guided by text-to-video priors. To generate cartoon-style and smooth motion, we first define B\'{e}zier curves over keypoints of the clipart image as a form of motion regularization. We then align the motion trajectories of the keypoints with the provided text prompt by optimizing the Video Score Distillation Sampling (VSDS) loss, which encodes adequate knowledge of natural motion within a pretrained text-to-video diffusion model. With a differentiable As-Rigid-As-Possible shape deformation algorithm, our method can be end-to-end optimized while maintaining deformation rigidity. Experimental results show that the proposed AniClipart consistently outperforms existing image-to-video generation models, in terms of text-video alignment, visual identity preservation, and motion consistency. Furthermore, we showcase the versatility of AniClipart by adapting it to generate a broader array of animation formats, such as layered animation, which allows topological changes.
Authors: Mengyuan Liu, Zhongbin Fang, Xia Li, Joachim M. Buhmann, Xiangtai Li, Chen Change Loy
Abstract: With the emergence of large-scale models trained on diverse datasets, in-context learning has emerged as a promising paradigm for multitasking, notably in natural language processing and image processing. However, its application in 3D point cloud tasks remains largely unexplored. In this work, we introduce Point-In-Context (PIC), a novel framework for 3D point cloud understanding via in-context learning. We address the technical challenge of effectively extending masked point modeling to 3D point clouds by introducing a Joint Sampling module and proposing a vanilla version of PIC called Point-In-Context-Generalist (PIC-G). PIC-G is designed as a generalist model for various 3D point cloud tasks, with inputs and outputs modeled as coordinates. In this paradigm, the challenging segmentation task is achieved by assigning label points with XYZ coordinates for each category; the final prediction is then chosen based on the label point closest to the predictions. To break the limitation by the fixed label-coordinate assignment, which has poor generalization upon novel classes, we propose two novel training strategies, In-Context Labeling and In-Context Enhancing, forming an extended version of PIC named Point-In-Context-Segmenter (PIC-S), targeting improving dynamic context labeling and model training. By utilizing dynamic in-context labels and extra in-context pairs, PIC-S achieves enhanced performance and generalization capability in and across part segmentation datasets. PIC is a general framework so that other tasks or datasets can be seamlessly introduced into our PIC through a unified data format. We conduct extensive experiments to validate the versatility and adaptability of our proposed methods in handling a wide range of tasks and segmenting multi-datasets. Our PIC-S is capable of generalizing unseen datasets and performing novel part segmentation by customizing prompts.
Authors: Hang Hua, Yunlong Tang, Chenliang Xu, Jiebo Luo
Abstract: Video summarization aims to create short, accurate, and cohesive summaries of longer videos. Despite the existence of various video summarization datasets, a notable limitation is their limited amount of source videos, which hampers the effective fine-tuning of advanced large vision-language models (VLMs). Additionally, most existing datasets are created for video-to-video summarization, overlooking the contemporary need for multimodal video content summarization. Recent efforts have been made to expand from unimodal to multimodal video summarization, categorizing the task into three sub-tasks based on the summary's modality: video-to-video (V2V), video-to-text (V2T), and a combination of video and text summarization (V2VT). However, the textual summaries in previous multimodal datasets are inadequate. To address these issues, we introduce Instruct-V2Xum, a cross-modal video summarization dataset featuring 30,000 diverse videos sourced from YouTube, with lengths ranging from 40 to 940 seconds and an average summarization ratio of 16.39\%. Each video summary in Instruct-V2Xum is paired with a textual summary that references specific frame indexes, facilitating the generation of aligned video and textual summaries. In addition, we propose a new video summarization framework named V2Xum-LLM. V2Xum-LLM, specifically V2Xum-LLaMA in this study, is the first framework that unifies different video summarization tasks into one large language model's (LLM) text decoder and achieves task-controllable video summarization with temporal prompts and task instructions. Experiments show that V2Xum-LLaMA outperforms strong baseline models on multiple video summarization tasks. Furthermore, we propose an enhanced evaluation metric for V2V and V2VT summarization tasks.
Authors: Julian Ost, Tanushree Banerjee, Mario Bijelic, Felix Heide
Abstract: Today, most methods for image understanding tasks rely on feed-forward neural networks. While this approach has allowed for empirical accuracy, efficiency, and task adaptation via fine-tuning, it also comes with fundamental disadvantages. Existing networks often struggle to generalize across different datasets, even on the same task. By design, these networks ultimately reason about high-dimensional scene features, which are challenging to analyze. This is true especially when attempting to predict 3D information based on 2D images. We propose to recast 3D multi-object tracking from RGB cameras as an \emph{Inverse Rendering (IR)} problem, by optimizing via a differentiable rendering pipeline over the latent space of pre-trained 3D object representations and retrieve the latents that best represent object instances in a given input image. To this end, we optimize an image loss over generative latent spaces that inherently disentangle shape and appearance properties. We investigate not only an alternate take on tracking but our method also enables examining the generated objects, reasoning about failure situations, and resolving ambiguous cases. We validate the generalization and scaling capabilities of our method by learning the generative prior exclusively from synthetic data and assessing camera-based 3D tracking on the nuScenes and Waymo datasets. Both these datasets are completely unseen to our method and do not require fine-tuning. Videos and code are available at https://light.princeton.edu/inverse-rendering-tracking/.
URLs: https://light.princeton.edu/inverse-rendering-tracking/.
Authors: Sina Sharifi, Taha Entesari, Bardia Safaei, Vishal M. Patel, Mahyar Fazlyab
Abstract: One of the challenges for neural networks in real-life applications is the overconfident errors these models make when the data is not from the original training distribution. Addressing this issue is known as Out-of-Distribution (OOD) detection. Many state-of-the-art OOD methods employ an auxiliary dataset as a surrogate for OOD data during training to achieve improved performance. However, these methods fail to fully exploit the local information embedded in the auxiliary dataset. In this work, we propose the idea of leveraging the information embedded in the gradient of the loss function during training to enable the network to not only learn a desired OOD score for each sample but also to exhibit similar behavior in a local neighborhood around each sample. We also develop a novel energy-based sampling method to allow the network to be exposed to more informative OOD samples during the training phase. This is especially important when the auxiliary dataset is large. We demonstrate the effectiveness of our method through extensive experiments on several OOD benchmarks, improving the existing state-of-the-art FPR95 by 4% on our ImageNet experiment. We further provide a theoretical analysis through the lens of certified robustness and Lipschitz analysis to showcase the theoretical foundation of our work. We will publicly release our code after the review process.
Authors: Xiaotang Gai, Chenyi Zhou, Jiaxiang Liu, Yang Feng, Jian Wu, Zuozhu Liu
Abstract: Medical Visual Question Answering (MedVQA), which offers language responses to image-based medical inquiries, represents a challenging task and significant advancement in healthcare. It assists medical experts to swiftly interpret medical images, thereby enabling faster and more accurate diagnoses. However, the model interpretability and transparency of existing MedVQA solutions are often limited, posing challenges in understanding their decision-making processes. To address this issue, we devise a semi-automated annotation process to streamlining data preparation and build new benchmark MedVQA datasets R-RAD and R-SLAKE. The R-RAD and R-SLAKE datasets provide intermediate medical decision-making rationales generated by multimodal large language models and human annotations for question-answering pairs in existing MedVQA datasets, i.e., VQA-RAD and SLAKE. Moreover, we design a novel framework which finetunes lightweight pretrained generative models by incorporating medical decision-making rationales into the training process. The framework includes three distinct strategies to generate decision outcomes and corresponding rationales, thereby clearly showcasing the medical decision-making process during reasoning. Extensive experiments demonstrate that our method can achieve an accuracy of 83.5% on R-RAD and 86.3% on R-SLAKE, significantly outperforming existing state-of-the-art baselines. Dataset and code will be released.
Authors: Th\'eo Gieruc, Marius K\"astingsch\"afer, Sebastian Bernhard, Mathieu Salzmann
Abstract: Current 3D reconstruction techniques struggle to infer unbounded scenes from a few images faithfully. Specifically, existing methods have high computational demands, require detailed pose information, and cannot reconstruct occluded regions reliably. We introduce 6Img-to-3D, an efficient, scalable transformer-based encoder-renderer method for single-shot image to 3D reconstruction. Our method outputs a 3D-consistent parameterized triplane from only six outward-facing input images for large-scale, unbounded outdoor driving scenarios. We take a step towards resolving existing shortcomings by combining contracted custom cross- and self-attention mechanisms for triplane parameterization, differentiable volume rendering, scene contraction, and image feature projection. We showcase that six surround-view vehicle images from a single timestamp without global pose information are enough to reconstruct 360$^{\circ}$ scenes during inference time, taking 395 ms. Our method allows, for example, rendering third-person images and birds-eye views. Our code is available at https://github.com/continental/6Img-to-3D, and more examples can be found at our website here https://6Img-to-3D.GitHub.io/.
URLs: https://github.com/continental/6Img-to-3D,, https://6Img-to-3D.GitHub.io/.
Authors: Isabella Liu, Hao Su, Xiaolong Wang
Abstract: Modern 3D engines and graphics pipelines require mesh as a memory-efficient representation, which allows efficient rendering, geometry processing, texture editing, and many other downstream operations. However, it is still highly difficult to obtain high-quality mesh in terms of structure and detail from monocular visual observations. The problem becomes even more challenging for dynamic scenes and objects. To this end, we introduce Dynamic Gaussians Mesh (DG-Mesh), a framework to reconstruct a high-fidelity and time-consistent mesh given a single monocular video. Our work leverages the recent advancement in 3D Gaussian Splatting to construct the mesh sequence with temporal consistency from a video. Building on top of this representation, DG-Mesh recovers high-quality meshes from the Gaussian points and can track the mesh vertices over time, which enables applications such as texture editing on dynamic objects. We introduce the Gaussian-Mesh Anchoring, which encourages evenly distributed Gaussians, resulting better mesh reconstruction through mesh-guided densification and pruning on the deformed Gaussians. By applying cycle-consistent deformation between the canonical and the deformed space, we can project the anchored Gaussian back to the canonical space and optimize Gaussians across all time frames. During the evaluation on different datasets, DG-Mesh provides significantly better mesh reconstruction and rendering than baselines.
Authors: Yotam Nitzan, Zongze Wu, Richard Zhang, Eli Shechtman, Daniel Cohen-Or, Taesung Park, Micha\"el Gharbi
Abstract: We introduce a novel diffusion transformer, LazyDiffusion, that generates partial image updates efficiently. Our approach targets interactive image editing applications in which, starting from a blank canvas or an image, a user specifies a sequence of localized image modifications using binary masks and text prompts. Our generator operates in two phases. First, a context encoder processes the current canvas and user mask to produce a compact global context tailored to the region to generate. Second, conditioned on this context, a diffusion-based transformer decoder synthesizes the masked pixels in a "lazy" fashion, i.e., it only generates the masked region. This contrasts with previous works that either regenerate the full canvas, wasting time and computation, or confine processing to a tight rectangular crop around the mask, ignoring the global image context altogether. Our decoder's runtime scales with the mask size, which is typically small, while our encoder introduces negligible overhead. We demonstrate that our approach is competitive with state-of-the-art inpainting methods in terms of quality and fidelity while providing a 10x speedup for typical user interactions, where the editing mask represents 10% of the image.
Authors: Yufei Ye, Abhinav Gupta, Kris Kitani, Shubham Tulsiani
Abstract: We propose G-HOP, a denoising diffusion based generative prior for hand-object interactions that allows modeling both the 3D object and a human hand, conditioned on the object category. To learn a 3D spatial diffusion model that can capture this joint distribution, we represent the human hand via a skeletal distance field to obtain a representation aligned with the (latent) signed distance field for the object. We show that this hand-object prior can then serve as generic guidance to facilitate other tasks like reconstruction from interaction clip and human grasp synthesis. We believe that our model, trained by aggregating seven diverse real-world interaction datasets spanning across 155 categories, represents a first approach that allows jointly generating both hand and object. Our empirical evaluations demonstrate the benefit of this joint prior in video-based reconstruction and human grasp synthesis, outperforming current task-specific baselines. Project website: https://judyye.github.io/ghop-www
Authors: Xinyue Wei, Kai Zhang, Sai Bi, Hao Tan, Fujun Luan, Valentin Deschaintre, Kalyan Sunkavalli, Hao Su, Zexiang Xu
Abstract: We propose MeshLRM, a novel LRM-based approach that can reconstruct a high-quality mesh from merely four input images in less than one second. Different from previous large reconstruction models (LRMs) that focus on NeRF-based reconstruction, MeshLRM incorporates differentiable mesh extraction and rendering within the LRM framework. This allows for end-to-end mesh reconstruction by fine-tuning a pre-trained NeRF LRM with mesh rendering. Moreover, we improve the LRM architecture by simplifying several complex designs in previous LRMs. MeshLRM's NeRF initialization is sequentially trained with low- and high-resolution images; this new LRM training strategy enables significantly faster convergence and thereby leads to better quality with less compute. Our approach achieves state-of-the-art mesh reconstruction from sparse-view inputs and also allows for many downstream applications, including text-to-3D and single-image-to-3D generation. Project page: https://sarahweiii.github.io/meshlrm/
Authors: Shengcao Cao, Jiuxiang Gu, Jason Kuen, Hao Tan, Ruiyi Zhang, Handong Zhao, Ani Nenkova, Liang-Yan Gui, Tong Sun, Yu-Xiong Wang
Abstract: Open-world entity segmentation, as an emerging computer vision task, aims at segmenting entities in images without being restricted by pre-defined classes, offering impressive generalization capabilities on unseen images and concepts. Despite its promise, existing entity segmentation methods like Segment Anything Model (SAM) rely heavily on costly expert annotators. This work presents Self-supervised Open-world Hierarchical Entity Segmentation (SOHES), a novel approach that eliminates the need for human annotations. SOHES operates in three phases: self-exploration, self-instruction, and self-correction. Given a pre-trained self-supervised representation, we produce abundant high-quality pseudo-labels through visual feature clustering. Then, we train a segmentation model on the pseudo-labels, and rectify the noises in pseudo-labels via a teacher-student mutual-learning procedure. Beyond segmenting entities, SOHES also captures their constituent parts, providing a hierarchical understanding of visual entities. Using raw images as the sole training data, our method achieves unprecedented performance in self-supervised open-world segmentation, marking a significant milestone towards high-quality open-world entity segmentation in the absence of human-annotated masks. Project page: https://SOHES.github.io.
URLs: https://SOHES.github.io.
Authors: Yiran Xu, Taesung Park, Richard Zhang, Yang Zhou, Eli Shechtman, Feng Liu, Jia-Bin Huang, Difan Liu
Abstract: Video super-resolution (VSR) approaches have shown impressive temporal consistency in upsampled videos. However, these approaches tend to generate blurrier results than their image counterparts as they are limited in their generative capability. This raises a fundamental question: can we extend the success of a generative image upsampler to the VSR task while preserving the temporal consistency? We introduce VideoGigaGAN, a new generative VSR model that can produce videos with high-frequency details and temporal consistency. VideoGigaGAN builds upon a large-scale image upsampler -- GigaGAN. Simply inflating GigaGAN to a video model by adding temporal modules produces severe temporal flickering. We identify several key issues and propose techniques that significantly improve the temporal consistency of upsampled videos. Our experiments show that, unlike previous VSR methods, VideoGigaGAN generates temporally consistent videos with more fine-grained appearance details. We validate the effectiveness of VideoGigaGAN by comparing it with state-of-the-art VSR models on public datasets and showcasing video results with $8\times$ super-resolution.
Authors: Junyu Xie, Charig Yang, Weidi Xie, Andrew Zisserman
Abstract: The objective of this paper is motion segmentation -- discovering and segmenting the moving objects in a video. This is a much studied area with numerous careful,and sometimes complex, approaches and training schemes including: self-supervised learning, learning from synthetic datasets, object-centric representations, amodal representations, and many more. Our interest in this paper is to determine if the Segment Anything model (SAM) can contribute to this task. We investigate two models for combining SAM with optical flow that harness the segmentation power of SAM with the ability of flow to discover and group moving objects. In the first model, we adapt SAM to take optical flow, rather than RGB, as an input. In the second, SAM takes RGB as an input, and flow is used as a segmentation prompt. These surprisingly simple methods, without any further modifications, outperform all previous approaches by a considerable margin in both single and multi-object benchmarks. We also extend these frame-level segmentations to sequence-level segmentations that maintain object identity. Again, this simple model outperforms previous methods on multiple video object segmentation benchmarks.
Authors: Xingyu Fu, Yushi Hu, Bangzheng Li, Yu Feng, Haoyu Wang, Xudong Lin, Dan Roth, Noah A. Smith, Wei-Chiu Ma, Ranjay Krishna
Abstract: We introduce Blink, a new benchmark for multimodal language models (LLMs) that focuses on core visual perception abilities not found in other evaluations. Most of the Blink tasks can be solved by humans "within a blink" (e.g., relative depth estimation, visual correspondence, forensics detection, and multi-view reasoning). However, we find these perception-demanding tasks cast significant challenges for current multimodal LLMs because they resist mediation through natural language. Blink reformats 14 classic computer vision tasks into 3,807 multiple-choice questions, paired with single or multiple images and visual prompting. While humans get 95.70% accuracy on average, Blink is surprisingly challenging for existing multimodal LLMs: even the best-performing GPT-4V and Gemini achieve accuracies of 51.26% and 45.72%, only 13.17% and 7.63% higher than random guessing, indicating that such perception abilities have not "emerged" yet in recent multimodal LLMs. Our analysis also highlights that specialist CV models could solve these problems much better, suggesting potential pathways for future improvements. We believe Blink will stimulate the community to help multimodal LLMs catch up with human-level visual perception.
Authors: Songwei Ge, Aniruddha Mahapatra, Gaurav Parmar, Jun-Yan Zhu, Jia-Bin Huang
Abstract: Fr\'echet Video Distance (FVD), a prominent metric for evaluating video generation models, is known to conflict with human perception occasionally. In this paper, we aim to explore the extent of FVD's bias toward per-frame quality over temporal realism and identify its sources. We first quantify the FVD's sensitivity to the temporal axis by decoupling the frame and motion quality and find that the FVD increases only slightly with large temporal corruption. We then analyze the generated videos and show that via careful sampling from a large set of generated videos that do not contain motions, one can drastically decrease FVD without improving the temporal quality. Both studies suggest FVD's bias towards the quality of individual frames. We further observe that the bias can be attributed to the features extracted from a supervised video classifier trained on the content-biased dataset. We show that FVD with features extracted from the recent large-scale self-supervised video models is less biased toward image quality. Finally, we revisit a few real-world examples to validate our hypothesis.
Authors: Shivvrat Arya, Yu Xiang, Vibhav Gogate
Abstract: We present a unified framework called deep dependency networks (DDNs) that combines dependency networks and deep learning architectures for multi-label classification, with a particular emphasis on image and video data. The primary advantage of dependency networks is their ease of training, in contrast to other probabilistic graphical models like Markov networks. In particular, when combined with deep learning architectures, they provide an intuitive, easy-to-use loss function for multi-label classification. A drawback of DDNs compared to Markov networks is their lack of advanced inference schemes, necessitating the use of Gibbs sampling. To address this challenge, we propose novel inference schemes based on local search and integer linear programming for computing the most likely assignment to the labels given observations. We evaluate our novel methods on three video datasets (Charades, TACoS, Wetlab) and three image datasets (MS-COCO, PASCAL VOC, NUS-WIDE), comparing their performance with (a) basic neural architectures and (b) neural architectures combined with Markov networks equipped with advanced inference and learning techniques. Our results demonstrate the superiority of our new DDN methods over the two competing approaches.
Authors: Weiming Zhi, Haozhan Tang, Tianyi Zhang, Matthew Johnson-Roberson
Abstract: Representing the environment is a central challenge in robotics, and is essential for effective decision-making. Traditionally, before capturing images with a manipulator-mounted camera, users need to calibrate the camera using a specific external marker, such as a checkerboard or AprilTag. However, recent advances in computer vision have led to the development of \emph{3D foundation models}. These are large, pre-trained neural networks that can establish fast and accurate multi-view correspondences with very few images, even in the absence of rich visual features. This paper advocates for the integration of 3D foundation models into scene representation approaches for robotic systems equipped with manipulator-mounted RGB cameras. Specifically, we propose the Joint Calibration and Representation (JCR) method. JCR uses RGB images, captured by a manipulator-mounted camera, to simultaneously construct an environmental representation and calibrate the camera relative to the robot's end-effector, in the absence of specific calibration markers. The resulting 3D environment representation is aligned with the robot's coordinate frame and maintains physically accurate scales. We demonstrate that JCR can build effective scene representations using a low-cost RGB camera attached to a manipulator, without prior calibration.
Authors: Santiago Cepeda, Roberto Romero, Daniel Garcia-Perez, Guillermo Blasco, Luigi Tommaso Luppino, Samuel Kuttner, Ignacio Arrese, Ole Solheim, Live Eikenes, Anna Karlberg, Angel Perez-Nunez, Trinidad Escudero, Roberto Hornero, Rosario Sarabia
Abstract: Accurately assessing tumor removal is paramount in the management of glioblastoma. We developed a pipeline using MRI scans and neural networks to segment tumor subregions and the surgical cavity in postoperative images. Our model excels in accurately classifying the extent of resection, offering a valuable tool for clinicians in assessing treatment effectiveness.
Authors: A. Ren\'e Geist, Jonas Frey, Mikel Zobro, Anna Levina, Georg Martius
Abstract: Many settings in machine learning require the selection of a rotation representation. However, choosing a suitable representation from the many available options is challenging. This paper acts as a survey and guide through rotation representations. We walk through their properties that harm or benefit deep learning with gradient-based optimization. By consolidating insights from rotation-based learning, we provide a comprehensive overview of learning functions with rotation representations. We provide guidance on selecting representations based on whether rotations are in the model's input or output and whether the data primarily comprises small angles.
Authors: Muheng Li, Xia Li, Sairos Safai, Damien Weber, Antony Lomax, Ye Zhang
Abstract: In recent advancements in proton therapy, MR-based treatment planning is gaining momentum to minimize additional radiation exposure compared to traditional CT-based methods. This transition highlights the critical need for accurate MR-to-CT image synthesis, which is essential for precise proton dose calculations. Our research introduces the Diffusion Schr\"odinger Bridge Models (DSBM), an innovative approach for high-quality MR-to-CT synthesis. DSBM learns the nonlinear diffusion processes between MR and CT data distributions. This method improves upon traditional diffusion models by initiating synthesis from the prior distribution rather than the Gaussian distribution, enhancing both generation quality and efficiency. We validated the effectiveness of DSBM on a head and neck cancer dataset, demonstrating its superiority over traditional image synthesis methods through both image-level and dosimetric-level evaluations. The effectiveness of DSBM in MR-based proton treatment planning highlights its potential as a valuable tool in various clinical scenarios.
Authors: MohammadHossein AskariHemmat, Ahmadreza Jeddi, Reyhane Askari Hemmat, Ivan Lazarevich, Alexander Hoffman, Sudhakar Sah, Ehsan Saboori, Yvon Savaria, Jean-Pierre David
Abstract: Quantization lowers memory usage, computational requirements, and latency by utilizing fewer bits to represent model weights and activations. In this work, we investigate the generalization properties of quantized neural networks, a characteristic that has received little attention despite its implications on model performance. In particular, first, we develop a theoretical model for quantization in neural networks and demonstrate how quantization functions as a form of regularization. Second, motivated by recent work connecting the sharpness of the loss landscape and generalization, we derive an approximate bound for the generalization of quantized models conditioned on the amount of quantization noise. We then validate our hypothesis by experimenting with over 2000 models trained on CIFAR-10, CIFAR-100, and ImageNet datasets on convolutional and transformer-based models.
Authors: Rachel (Lei), Chen, Wenjia Zheng, Sandeep Jalui, Pavan Suri, Jun Zeng
Abstract: With the advancements in 3D printing technologies, it is extremely important that the quality of 3D printed objects, and dimensional accuracies should meet the customer's specifications. Various factors during metal printing affect the printed parts' quality, including the power quality, the printing stage parameters, the print part's location inside the print bed, the curing stage parameters, and the metal sintering process. With the large data gathered from HP's MetJet printing process, AI techniques can be used to analyze, learn, and effectively infer the printed part quality metrics, as well as assist in improving the print yield. In-situ thermal sensing data captured by printer-installed thermal sensors contains the part thermal signature of fusing layers. Such part thermal signature contains a convoluted impact from various factors. In this paper, we use a multimodal thermal encoder network to fuse data of a different nature including the video data vectorized printer control data, and exact part thermal signatures with a trained encoder-decoder module. We explored the data fusing techniques and stages for data fusing, the optimized end-to-end model architecture indicates an improved part quality prediction accuracy.
Authors: Marzi Heidari, Hanping Zhang, Yuhong Guo
Abstract: In this paper, we present a novel approach termed Prompt-Driven Feature Diffusion (PDFD) within a semi-supervised learning framework for Open World Semi-Supervised Learning (OW-SSL). At its core, PDFD deploys an efficient feature-level diffusion model with the guidance of class-specific prompts to support discriminative feature representation learning and feature generation, tackling the challenge of the non-availability of labeled data for unseen classes in OW-SSL. In particular, PDFD utilizes class prototypes as prompts in the diffusion model, leveraging their class-discriminative and semantic generalization ability to condition and guide the diffusion process across all the seen and unseen classes. Furthermore, PDFD incorporates a class-conditional adversarial loss for diffusion model training, ensuring that the features generated via the diffusion process can be discriminatively aligned with the class-conditional features of the real data. Additionally, the class prototypes of the unseen classes are computed using only unlabeled instances with confident predictions within a semi-supervised learning framework. We conduct extensive experiments to evaluate the proposed PDFD. The empirical results show PDFD exhibits remarkable performance enhancements over many state-of-the-art existing methods.
Authors: Sonit Singh
Abstract: Medical imaging has been used for diagnosis of various conditions, making it one of the most powerful resources for effective patient care. Due to widespread availability, low cost, and low radiation, chest X-ray is one of the most sought after radiology examination for the diagnosis of various thoracic diseases. Due to advancements in medical imaging technologies and increasing patient load, current radiology workflow faces various challenges including increasing backlogs, working long hours, and increase in diagnostic errors. An automated computer-aided diagnosis system that can interpret chest X-rays to augment radiologists by providing actionable insights has potential to provide second opinion to radiologists, highlight relevant regions in the image, in turn expediting clinical workflow, reducing diagnostic errors, and improving patient care. In this study, we applied a novel architecture augmenting the DenseNet121 Convolutional Neural Network (CNN) with multi-head self-attention mechanism using transformer, namely SA-DenseNet121, that can identify multiple thoracic diseases in chest X-rays. We conducted experiments on four of the largest chest X-ray datasets, namely, ChestX-ray14, CheXpert, MIMIC-CXR-JPG, and IU-CXR. Experimental results in terms of area under the receiver operating characteristics (AUC-ROC) shows that augmenting CNN with self-attention has potential in diagnosing different thoracic diseases from chest X-rays. The proposed methodology has the potential to support the reading workflow, improve efficiency, and reduce diagnostic errors.
Authors: Lixing Tan, Shuang Song, Kangneng Zhou, Chengbo Duan, Lanying Wang, Huayang Ren, Linlin Liu, Wei Zhang, Ruoxiu Xiao
Abstract: X-ray images play a vital role in the intraoperative processes due to their high resolution and fast imaging speed and greatly promote the subsequent segmentation, registration and reconstruction. However, over-dosed X-rays superimpose potential risks to human health to some extent. Data-driven algorithms from volume scans to X-ray images are restricted by the scarcity of paired X-ray and volume data. Existing methods are mainly realized by modelling the whole X-ray imaging procedure. In this study, we propose a learning-based approach termed CT2X-GAN to synthesize the X-ray images in an end-to-end manner using the content and style disentanglement from three different image domains. Our method decouples the anatomical structure information from CT scans and style information from unpaired real X-ray images/ digital reconstructed radiography (DRR) images via a series of decoupling encoders. Additionally, we introduce a novel consistency regularization term to improve the stylistic resemblance between synthesized X-ray images and real X-ray images. Meanwhile, we also impose a supervised process by computing the similarity of computed real DRR and synthesized DRR images. We further develop a pose attention module to fully strengthen the comprehensive information in the decoupled content code from CT scans, facilitating high-quality multi-view image synthesis in the lower 2D space. Extensive experiments were conducted on the publicly available CTSpine1K dataset and achieved 97.8350, 0.0842 and 3.0938 in terms of FID, KID and defined user-scored X-ray similarity, respectively. In comparison with 3D-aware methods ($\pi$-GAN, EG3D), CT2X-GAN is superior in improving the synthesis quality and realistic to the real X-ray images.
Authors: Thibault Castells, Hyoung-Kyu Song, Tairen Piao, Shinkook Choi, Bo-Kyeong Kim, Hanyoung Yim, Changgwun Lee, Jae Gon Kim, Tae-Ho Kim
Abstract: The intensive computational burden of Stable Diffusion (SD) for text-to-image generation poses a significant hurdle for its practical application. To tackle this challenge, recent research focuses on methods to reduce sampling steps, such as Latent Consistency Model (LCM), and on employing architectural optimizations, including pruning and knowledge distillation. Diverging from existing approaches, we uniquely start with a compact SD variant, BK-SDM. We observe that directly applying LCM to BK-SDM with commonly used crawled datasets yields unsatisfactory results. It leads us to develop two strategies: (1) leveraging high-quality image-text pairs from leading generative models and (2) designing an advanced distillation process tailored for LCM. Through our thorough exploration of quantization, profiling, and on-device deployment, we achieve rapid generation of photo-realistic, text-aligned images in just two steps, with latency under one second on resource-limited edge devices.
Authors: Walid Abdullah Al, Il Dong Yun, Yun Jung Bae
Abstract: Dopamine transporter (DAT) imaging is commonly used for monitoring Parkinson's disease (PD), where striatal DAT uptake amount is computed to assess PD severity. However, DAT imaging has a high cost and the risk of radiance exposure and is not available in general clinics. Recently, MRI patch of the nigral region has been proposed as a safer and easier alternative. This paper proposes a symmetric regressor for predicting the DAT uptake amount from the nigral MRI patch. Acknowledging the symmetry between the right and left nigrae, the proposed regressor incorporates a paired input-output model that simultaneously predicts the DAT uptake amounts for both the right and left striata. Moreover, it employs a symmetric loss that imposes a constraint on the difference between right-to-left predictions, resembling the high correlation in DAT uptake amounts in the two lateral sides. Additionally, we propose a symmetric Monte-Carlo (MC) dropout method for providing a fruitful uncertainty estimate of the DAT uptake prediction, which utilizes the above symmetry. We evaluated the proposed approach on 734 nigral patches, which demonstrated significantly improved performance of the symmetric regressor compared with the standard regressors while giving better explainability and feature representation. The symmetric MC dropout also gave precise uncertainty ranges with a high probability of including the true DAT uptake amounts within the range.
Authors: Thibault Castells, Hyoung-Kyu Song, Bo-Kyeong Kim, Shinkook Choi
Abstract: Latent Diffusion Models (LDMs) have emerged as powerful generative models, known for delivering remarkable results under constrained computational resources. However, deploying LDMs on resource-limited devices remains a complex issue, presenting challenges such as memory consumption and inference speed. To address this issue, we introduce LD-Pruner, a novel performance-preserving structured pruning method for compressing LDMs. Traditional pruning methods for deep neural networks are not tailored to the unique characteristics of LDMs, such as the high computational cost of training and the absence of a fast, straightforward and task-agnostic method for evaluating model performance. Our method tackles these challenges by leveraging the latent space during the pruning process, enabling us to effectively quantify the impact of pruning on model performance, independently of the task at hand. This targeted pruning of components with minimal impact on the output allows for faster convergence during training, as the model has less information to re-learn, thereby addressing the high computational cost of training. Consequently, our approach achieves a compressed model that offers improved inference speed and reduced parameter count, while maintaining minimal performance degradation. We demonstrate the effectiveness of our approach on three different tasks: text-to-image (T2I) generation, Unconditional Image Generation (UIG) and Unconditional Audio Generation (UAG). Notably, we reduce the inference time of Stable Diffusion (SD) by 34.9% while simultaneously improving its FID by 5.2% on MS-COCO T2I benchmark. This work paves the way for more efficient pruning methods for LDMs, enhancing their applicability.
Authors: Xiao Wang, Ke Tang, Xingyuan Dai, Jintao Xu, Quancheng Du, Rui Ai, Yuxiao Wang, Weihao Gu
Abstract: In public roads, autonomous vehicles (AVs) face the challenge of frequent interactions with human-driven vehicles (HDVs), which render uncertain driving behavior due to varying social characteristics among humans. To effectively assess the risks prevailing in the vicinity of AVs in social interactive traffic scenarios and achieve safe autonomous driving, this article proposes a social-suitable and safety-sensitive trajectory planning (S4TP) framework. Specifically, S4TP integrates the Social-Aware Trajectory Prediction (SATP) and Social-Aware Driving Risk Field (SADRF) modules. SATP utilizes Transformers to effectively encode the driving scene and incorporates an AV's planned trajectory during the prediction decoding process. SADRF assesses the expected surrounding risk degrees during AVs-HDVs interactions, each with different social characteristics, visualized as two-dimensional heat maps centered on the AV. SADRF models the driving intentions of the surrounding HDVs and predicts trajectories based on the representation of vehicular interactions. S4TP employs an optimization-based approach for motion planning, utilizing the predicted HDVs'trajectories as input. With the integration of SADRF, S4TP executes real-time online optimization of the planned trajectory of AV within lowrisk regions, thus improving the safety and the interpretability of the planned trajectory. We have conducted comprehensive tests of the proposed method using the SMARTS simulator. Experimental results in complex social scenarios, such as unprotected left turn intersections, merging, cruising, and overtaking, validate the superiority of our proposed S4TP in terms of safety and rationality. S4TP achieves a pass rate of 100% across all scenarios, surpassing the current state-of-the-art methods Fanta of 98.25% and Predictive-Decision of 94.75%.
Authors: Shijie Fang, Qianhan Feng, Tong Lin
Abstract: Despite the progress of Semi-supervised Learning (SSL), existing methods fail to utilize unlabeled data effectively and efficiently. Many pseudo-label-based methods select unlabeled examples based on inaccurate confidence scores from the classifier. Most prior work also uses all available unlabeled data without pruning, making it difficult to handle large amounts of unlabeled data. To address these issues, we propose two methods: Variational Confidence Calibration (VCC) and Influence-Function-based Unlabeled Sample Elimination (INFUSE). VCC is an universal plugin for SSL confidence calibration, using a variational autoencoder to select more accurate pseudo labels based on three types of consistency scores. INFUSE is a data pruning method that constructs a core dataset of unlabeled examples under SSL. Our methods are effective in multiple datasets and settings, reducing classification errors rates and saving training time. Together, VCC-INFUSE reduces the error rate of FlexMatch on the CIFAR-100 dataset by 1.08% while saving nearly half of the training time.
Authors: Chao Zhou, Huishuai Zhang, Jiang Bian, Weiming Zhang, Nenghai Yu
Abstract: This paper addresses the contentious issue of copyright infringement in images generated by text-to-image models, sparking debates among AI developers, content creators, and legal entities. State-of-the-art models create high-quality content without crediting original creators, causing concern in the artistic community. To mitigate this, we propose the \copyright Plug-in Authorization framework, introducing three operations: addition, extraction, and combination. Addition involves training a \copyright plug-in for specific copyright, facilitating proper credit attribution. Extraction allows creators to reclaim copyright from infringing models, and combination enables users to merge different \copyright plug-ins. These operations act as permits, incentivizing fair use and providing flexibility in authorization. We present innovative approaches,"Reverse LoRA" for extraction and "EasyMerge" for seamless combination. Experiments in artist-style replication and cartoon IP recreation demonstrate \copyright plug-ins' effectiveness, offering a valuable solution for human copyright protection in the age of generative AIs.
Authors: Robert J\"ochl, Andreas Uhl
Abstract: The goal of temporal image forensic is to approximate the age of a digital image relative to images from the same device. Usually, this is based on traces left during the image acquisition pipeline. For example, several methods exist that exploit the presence of in-field sensor defects for this purpose. In addition to these 'classical' methods, there is also an approach in which a Convolutional Neural Network (CNN) is trained to approximate the image age. One advantage of a CNN is that it independently learns the age features used. This would make it possible to exploit other (different) age traces in addition to the known ones (i.e., in-field sensor defects). In a previous work, we have shown that the presence of strong in-field sensor defects is irrelevant for a CNN to predict the age class. Based on this observation, the question arises how device (in)dependent the learned features are. In this work, we empirically asses this by training a network on images from a single device and then apply the trained model to images from different devices. This evaluation is performed on 14 different devices, including 10 devices from the publicly available 'Northumbria Temporal Image Forensics' database. These 10 different devices are based on five different device pairs (i.e., with the identical camera model).
Authors: Jinwu Wang, Wei Mao, Miaomiao Liu
Abstract: In this paper, we introduce a MusIc conditioned 3D Dance GEneraTion model, named MIDGET based on Dance motion Vector Quantised Variational AutoEncoder (VQ-VAE) model and Motion Generative Pre-Training (GPT) model to generate vibrant and highquality dances that match the music rhythm. To tackle challenges in the field, we introduce three new components: 1) a pre-trained memory codebook based on the Motion VQ-VAE model to store different human pose codes, 2) employing Motion GPT model to generate pose codes with music and motion Encoders, 3) a simple framework for music feature extraction. We compare with existing state-of-the-art models and perform ablation experiments on AIST++, the largest publicly available music-dance dataset. Experiments demonstrate that our proposed framework achieves state-of-the-art performance on motion quality and its alignment with the music.
Authors: Naibo Wang, Yuchen Deng, Wenjie Feng, Shichen Fan, Jianwei Yin, See-Kiong Ng
Abstract: Traditional federated learning mainly focuses on parallel settings (PFL), which can suffer significant communication and computation costs. In contrast, one-shot and sequential federated learning (SFL) have emerged as innovative paradigms to alleviate these costs. However, the issue of non-IID (Independent and Identically Distributed) data persists as a significant challenge in one-shot and SFL settings, exacerbated by the restricted communication between clients. In this paper, we improve the one-shot sequential federated learning for non-IID data by proposing a local model diversity-enhancing strategy. Specifically, to leverage the potential of local model diversity for improving model performance, we introduce a local model pool for each client that comprises diverse models generated during local training, and propose two distance measurements to further enhance the model diversity and mitigate the effect of non-IID data. Consequently, our proposed framework can improve the global model performance while maintaining low communication costs. Extensive experiments demonstrate that our method exhibits superior performance to existing one-shot PFL methods and achieves better accuracy compared with state-of-the-art one-shot SFL methods on both label-skew and domain-shift tasks (e.g., 6%+ accuracy improvement on the CIFAR-10 dataset).
Authors: Mary Aiyetigbo, Alexander Korte, Ethan Anderson, Reda Chalhoub, Peter Kalivas, Feng Luo, Nianyi Li
Abstract: In this paper, we introduce a novel unsupervised network to denoise microscopy videos featured by image sequences captured by a fixed location microscopy camera. Specifically, we propose a DeepTemporal Interpolation method, leveraging a temporal signal filter integrated into the bottom CNN layers, to restore microscopy videos corrupted by unknown noise types. Our unsupervised denoising architecture is distinguished by its ability to adapt to multiple noise conditions without the need for pre-existing noise distribution knowledge, addressing a significant challenge in real-world medical applications. Furthermore, we evaluate our denoising framework using both real microscopy recordings and simulated data, validating our outperforming video denoising performance across a broad spectrum of noise scenarios. Extensive experiments demonstrate that our unsupervised model consistently outperforms state-of-the-art supervised and unsupervised video denoising techniques, proving especially effective for microscopy videos.
Authors: Luciana Trinkaus Menon, Luiz Carlos Ribeiro Neduziak, Jean Paul Barddal, Alessandro Lameiras Koerich, Alceu de Souza Britto Jr
Abstract: The study of human emotions, traditionally a cornerstone in fields like psychology and neuroscience, has been profoundly impacted by the advent of artificial intelligence (AI). Multiple channels, such as speech (voice) and facial expressions (image), are crucial in understanding human emotions. However, AI's journey in multimodal emotion recognition (MER) is marked by substantial technical challenges. One significant hurdle is how AI models manage the absence of a particular modality - a frequent occurrence in real-world situations. This study's central focus is assessing the performance and resilience of two strategies when confronted with the lack of one modality: a novel multimodal dynamic modality and view selection and a cross-attention mechanism. Results on the RECOLA dataset show that dynamic selection-based methods are a promising approach for MER. In the missing modalities scenarios, all dynamic selection-based methods outperformed the baseline. The study concludes by emphasizing the intricate interplay between audio and video modalities in emotion prediction, showcasing the adaptability of dynamic selection methods in handling missing modalities.
Authors: Spencer Carmichael, Rahul Agrawal, Ram Vasudevan, Katherine A. Skinner
Abstract: Recognizing places from an opposing viewpoint during a return trip is a common experience for human drivers. However, the analogous robotics capability, visual place recognition (VPR) with limited field of view cameras under 180 degree rotations, has proven to be challenging to achieve. To address this problem, this paper presents Same Place Opposing Trajectory (SPOT), a technique for opposing viewpoint VPR that relies exclusively on structure estimated through stereo visual odometry (VO). The method extends recent advances in lidar descriptors and utilizes a novel double (similar and opposing) distance matrix sequence matching method. We evaluate SPOT on a publicly available dataset with 6.7-7.6 km routes driven in similar and opposing directions under various lighting conditions. The proposed algorithm demonstrates remarkable improvement over the state-of-the-art, achieving up to 91.7% recall at 100% precision in opposing viewpoint cases, while requiring less storage than all baselines tested and running faster than all but one. Moreover, the proposed method assumes no a priori knowledge of whether the viewpoint is similar or opposing, and also demonstrates competitive performance in similar viewpoint cases.
Authors: Yinzhu Jin, Matthew B. Dwyer, P. Thomas Fletcher
Abstract: This paper introduces a new technique to measure the feature dependency of neural network models. The motivation is to better understand a model by querying whether it is using information from human-understandable features, e.g., anatomical shape, volume, or image texture. Our method is based on the principle that if a model is dependent on a feature, then removal of that feature should significantly harm its performance. A targeted feature is "removed" by collapsing the dimension in the data distribution that corresponds to that feature. We perform this by moving data points along the feature dimension to a baseline feature value while staying on the data manifold, as estimated by a deep generative model. Then we observe how the model's performance changes on the modified test data set, with the target feature dimension removed. We test our method on deep neural network models trained on synthetic image data with known ground truth, an Alzheimer's disease prediction task using MRI and hippocampus segmentations from the OASIS-3 dataset, and a cell nuclei classification task using the Lizard dataset.
Authors: Aitor Ormazabal, Che Zheng, Cyprien de Masson d'Autume, Dani Yogatama, Deyu Fu, Donovan Ong, Eric Chen, Eugenie Lamprecht, Hai Pham, Isaac Ong, Kaloyan Aleksiev, Lei Li, Matthew Henderson, Max Bain, Mikel Artetxe, Nishant Relan, Piotr Padlewski, Qi Liu, Ren Chen, Samuel Phua, Yazheng Yang, Yi Tay, Yuqi Wang, Zhongkai Zhu, Zhihui Xie
Abstract: We introduce Reka Core, Flash, and Edge, a series of powerful multimodal language models trained from scratch by Reka. Reka models are able to process and reason with text, images, video, and audio inputs. This technical report discusses details of training some of these models and provides comprehensive evaluation results. We show that Reka Edge and Reka Flash are not only state-of-the-art but also outperform many much larger models, delivering outsized values for their respective compute class. Meanwhile, our most capable and largest model, Reka Core, approaches the best frontier models on both automatic evaluations and blind human evaluations. On image question answering benchmarks (e.g. MMMU, VQAv2), Core performs competitively to GPT4-V. Meanwhile, on multimodal chat, Core ranks as the second most preferred model under a blind third-party human evaluation setup, outperforming other models such as Claude 3 Opus. On text benchmarks, Core not only performs competitively to other frontier models on a set of well-established benchmarks (e.g. MMLU, GSM8K) but also outperforms GPT4-0613 on human evaluation. On video question answering (Perception-Test), Core outperforms Gemini Ultra. Models are shipped in production at http://chat.reka.ai . A showcase of non cherry picked qualitative examples can also be found at http://showcase.reka.ai .
Authors: Minjung Shin, Seongho Choi, Yu-Jung Heo, Minsu Lee, Byoung-Tak Zhang, Jeh-Kwang Ryu
Abstract: We introduce CogME, a cognition-inspired, multi-dimensional evaluation metric designed for AI models focusing on story understanding. CogME is a framework grounded in human thinking strategies and story elements that involve story understanding. With a specific breakdown of the questions, this approach provides a nuanced assessment revealing not only AI models' particular strengths and weaknesses but also the characteristics of the benchmark dataset. Our case study with the DramaQA dataset demonstrates a refined analysis of the model and the benchmark dataset. We argue the need for metrics based on understanding the nature of tasks and designed to align closely with human cognitive processes. This approach provides insights beyond traditional overall scores and paves the way for more sophisticated AI development targeting higher cognitive functions.
Authors: Lorenzo Luzi, Paul M Mayer, Josue Casco-Rodriguez, Ali Siahkoohi, Richard G. Baraniuk
Abstract: The inference stage of diffusion models can be seen as running a reverse-time diffusion stochastic differential equation, where samples from a Gaussian latent distribution are transformed into samples from a target distribution that usually reside on a low-dimensional manifold, e.g., an image manifold. The intermediate values between the initial latent space and the image manifold can be interpreted as noisy images, with the amount of noise determined by the forward diffusion process noise schedule. We utilize this interpretation to present Boomerang, an approach for local sampling of image manifolds. As implied by its name, Boomerang local sampling involves adding noise to an input image, moving it closer to the latent space, and then mapping it back to the image manifold through a partial reverse diffusion process. Thus, Boomerang generates images on the manifold that are ``similar,'' but nonidentical, to the original input image. We can control the proximity of the generated images to the original by adjusting the amount of noise added. Furthermore, due to the stochastic nature of the reverse diffusion process in Boomerang, the generated images display a certain degree of stochasticity, allowing us to obtain local samples from the manifold without encountering any duplicates. Boomerang offers the flexibility to work seamlessly with any pretrained diffusion model, such as Stable Diffusion, without necessitating any adjustments to the reverse diffusion process. We present three applications for Boomerang. First, we provide a framework for constructing privacy-preserving datasets having controllable degrees of anonymity. Second, we show that using Boomerang for data augmentation increases generalization performance and outperforms state-of-the-art synthetic data augmentation. Lastly, we introduce a perceptual image enhancement framework, which enables resolution enhancement.
Authors: Mir Rayat Imtiaz Hossain, Leonid Sigal, James J. Little
Abstract: Recent advances in pixel-level tasks (e.g. segmentation) illustrate the benefit of of long-range interactions between aggregated region-based representations that can enhance local features. However, such aggregated representations, often in the form of attention, fail to model the underlying semantics of the scene (e.g. individual objects and, by extension, their interactions). In this work, we address the issue by proposing a component that learns to project image features into latent representations and reason between them using a transformer encoder to generate contextualized and scene-consistent representations which are fused with original image features. Our design encourages the latent regions to represent semantic concepts by ensuring that the activated regions are spatially disjoint and the union of such regions corresponds to a connected object segment. The proposed semantic global reasoning (SGR) component is end-to-end trainable and can be easily added to a wide variety of backbones (CNN or transformer-based) and segmentation heads (per-pixel or mask classification) to consistently improve the segmentation results on different datasets. In addition, our latent tokens are semantically interpretable and diverse and provide a rich set of features that can be transferred to downstream tasks like object detection and segmentation, with improved performance. Furthermore, we also proposed metrics to quantify the semantics of latent tokens at both class \& instance level.
Authors: Matheus Viana da Silva, Nat\'alia de Carvalho Santos, Julie Ouellette, Baptiste Lacoste, Cesar Henrique Comin
Abstract: Creating a dataset for training supervised machine learning algorithms can be a demanding task. This is especially true for medical image segmentation since one or more specialists are usually required for image annotation, and creating ground truth labels for just a single image can take up to several hours. In addition, it is paramount that the annotated samples represent well the different conditions that might affect the imaged tissues as well as possible changes in the image acquisition process. This can only be achieved by considering samples that are typical in the dataset as well as atypical, or even outlier, samples. We introduce VessMAP, a heterogeneous blood vessel segmentation dataset acquired by carefully sampling relevant images from a larger non-annotated dataset. A methodology was developed to select both prototypical and atypical samples from the base dataset, thus defining an assorted set of images that can be used for measuring the performance of segmentation algorithms on samples that are highly distinct from each other. To demonstrate the potential of the new dataset, we show that the validation performance of a neural network changes significantly depending on the splits used for training the network.
Authors: Yanbiao Ma, Licheng Jiao, Fang Liu, Maoji Wen, Lingling Li, Wenping Ma, Shuyuan Yang, Xu Liu, Puhua Chen
Abstract: To address the challenges of long-tailed classification, researchers have proposed several approaches to reduce model bias, most of which assume that classes with few samples are weak classes. However, recent studies have shown that tail classes are not always hard to learn, and model bias has been observed on sample-balanced datasets, suggesting the existence of other factors that affect model bias. In this work, we first establish a geometric perspective for analyzing model fairness and then systematically propose a series of geometric measurements for perceptual manifolds in deep neural networks. Subsequently, we comprehensively explore the effect of the geometric characteristics of perceptual manifolds on classification difficulty and how learning shapes the geometric characteristics of perceptual manifolds. An unanticipated finding is that the correlation between the class accuracy and the separation degree of perceptual manifolds gradually decreases during training, while the negative correlation with the curvature gradually increases, implying that curvature imbalance leads to model bias.Building upon these observations, we propose curvature regularization to facilitate the model to learn curvature-balanced and flatter perceptual manifolds. Evaluations on multiple long-tailed and non-long-tailed datasets show the excellent performance and exciting generality of our approach, especially in achieving significant performance improvements based on current state-of-the-art techniques. Our work opens up a geometric analysis perspective on model bias and reminds researchers to pay attention to model bias on non-long-tailed and even sample-balanced datasets.
Authors: Bohan Li, Yasheng Sun, Zhujin Liang, Dalong Du, Zhuanghui Zhang, Xiaofeng Wang, Yunnan Wang, Xin Jin, Wenjun Zeng
Abstract: 3D semantic scene completion (SSC) is an ill-posed perception task that requires inferring a dense 3D scene from limited observations. Previous camera-based methods struggle to predict accurate semantic scenes due to inherent geometric ambiguity and incomplete observations. In this paper, we resort to stereo matching technique and bird's-eye-view (BEV) representation learning to address such issues in SSC. Complementary to each other, stereo matching mitigates geometric ambiguity with epipolar constraint while BEV representation enhances the hallucination ability for invisible regions with global semantic context. However, due to the inherent representation gap between stereo geometry and BEV features, it is non-trivial to bridge them for dense prediction task of SSC. Therefore, we further develop a unified occupancy-based framework dubbed BRGScene, which effectively bridges these two representations with dense 3D volumes for reliable semantic scene completion. Specifically, we design a novel Mutual Interactive Ensemble (MIE) block for pixel-level reliable aggregation of stereo geometry and BEV features. Within the MIE block, a Bi-directional Reliable Interaction (BRI) module, enhanced with confidence re-weighting, is employed to encourage fine-grained interaction through mutual guidance. Besides, a Dual Volume Ensemble (DVE) module is introduced to facilitate complementary aggregation through channel-wise recalibration and multi-group voting. Our method outperforms all published camera-based methods on SemanticKITTI for semantic scene completion. Our code is available on \url{https://github.com/Arlo0o/StereoScene}.
Authors: Asude Aydin, Mathias Gehrig, Daniel Gehrig, Davide Scaramuzza
Abstract: Spiking Neural Networks (SNN) are a class of bio-inspired neural networks that promise to bring low-power and low-latency inference to edge devices through asynchronous and sparse processing. However, being temporal models, SNNs depend heavily on expressive states to generate predictions on par with classical artificial neural networks (ANNs). These states converge only after long transient periods, and quickly decay without input data, leading to higher latency, power consumption, and lower accuracy. This work addresses this issue by initializing the state with an auxiliary ANN running at a low rate. The SNN then uses the state to generate predictions with high temporal resolution until the next initialization phase. Our hybrid ANN-SNN model thus combines the best of both worlds: It does not suffer from long state transients and state decay thanks to the ANN, and can generate predictions with high temporal resolution, low latency, and low power thanks to the SNN. We show for the task of event-based 2D and 3D human pose estimation that our method consumes 88% less power with only a 4% decrease in performance compared to its fully ANN counterparts when run at the same inference rate. Moreover, when compared to SNNs, our method achieves a 74% lower error. This research thus provides a new understanding of how ANNs and SNNs can be used to maximize their respective benefits.
Authors: Mohsen Jenadeleh, Johannes Zagermann, Harald Reiterer, Ulf-Dietrich Reips, Raouf Hamzaoui, Dietmar Saupe
Abstract: In image quality assessment, a collective visual quality score for an image or video is obtained from the individual ratings of many subjects. One commonly used format for these experiments is the two-alternative forced choice method. Two stimuli with the same content but differing visual quality are presented sequentially or side-by-side. Subjects are asked to select the one of better quality, and when uncertain, they are required to guess. The relaxed alternative forced choice format aims to reduce the cognitive load and the noise in the responses due to the guessing by providing a third response option, namely, ``not sure''. This work presents a large and comprehensive crowdsourcing experiment to compare these two response formats: the one with the ``not sure'' option and the one without it. To provide unambiguous ground truth for quality evaluation, subjects were shown pairs of images with differing numbers of dots and asked each time to choose the one with more dots. Our crowdsourcing study involved 254 participants and was conducted using a within-subject design. Each participant was asked to respond to 40 pair comparisons with and without the ``not sure'' response option and completed a questionnaire to evaluate their cognitive load for each testing condition. The experimental results show that the inclusion of the ``not sure'' response option in the forced choice method reduced mental load and led to models with better data fit and correspondence to ground truth. We also tested for the equivalence of the models and found that they were different. The dataset is available at http://database.mmsp-kn.de/cogvqa-database.html.
Authors: Haram Choi, Cheolwoong Na, Jihyeon Oh, Seungjae Lee, Jinseop Kim, Subeen Choe, Jeongmin Lee, Taehoon Kim, Jihoon Yang
Abstract: Although many recent works have made advancements in the image restoration (IR) field, they often suffer from an excessive number of parameters. Another issue is that most Transformer-based IR methods focus only on either local or global features, leading to limited receptive fields or deficient parameter issues. To address these problems, we propose a lightweight IR network, Reciprocal Attention Mixing Transformer (RAMiT). It employs our proposed dimensional reciprocal attention mixing Transformer (D-RAMiT) blocks, which compute bi-dimensional (spatial and channel) self-attentions in parallel with different numbers of multi-heads. The bi-dimensional attentions help each other to complement their counterpart's drawbacks and are then mixed. Additionally, we introduce a hierarchical reciprocal attention mixing (H-RAMi) layer that compensates for pixel-level information losses and utilizes semantic information while maintaining an efficient hierarchical structure. Furthermore, we revisit and modify MobileNet V1 and V2 to attach efficient convolutions to our proposed components. The experimental results demonstrate that RAMiT achieves state-of-the-art performance on multiple lightweight IR tasks, including super-resolution, color denoising, grayscale denoising, low-light enhancement, and deraining. Codes are available at https://github.com/rami0205/RAMiT.
Authors: Yichun Shi, Peng Wang, Jianglong Ye, Mai Long, Kejie Li, Xiao Yang
Abstract: We introduce MVDream, a diffusion model that is able to generate consistent multi-view images from a given text prompt. Learning from both 2D and 3D data, a multi-view diffusion model can achieve the generalizability of 2D diffusion models and the consistency of 3D renderings. We demonstrate that such a multi-view diffusion model is implicitly a generalizable 3D prior agnostic to 3D representations. It can be applied to 3D generation via Score Distillation Sampling, significantly enhancing the consistency and stability of existing 2D-lifting methods. It can also learn new concepts from a few 2D examples, akin to DreamBooth, but for 3D generation.
Authors: Henry Hengyuan Zhao, Pichao Wang, Yuyang Zhao, Hao Luo, Fan Wang, Mike Zheng Shou
Abstract: Pre-trained vision transformers have strong representation benefits to various downstream tasks. Recently, many parameter-efficient fine-tuning (PEFT) methods have been proposed, and their experiments demonstrate that tuning only 1% of extra parameters could surpass full fine-tuning in low-data resource scenarios. However, these methods overlook the task-specific information when fine-tuning diverse downstream tasks. In this paper, we propose a simple yet effective method called "Salient Channel Tuning" (SCT) to leverage the task-specific information by forwarding the model with the task images to select partial channels in a feature map that enables us to tune only 1/8 channels leading to significantly lower parameter costs. Experiments outperform full fine-tuning on 18 out of 19 tasks in the VTAB-1K benchmark by adding only 0.11M parameters of the ViT-B, which is 780x fewer than its full fine-tuning counterpart. Furthermore, experiments on domain generalization and few-shot learning surpass other PEFT methods with lower parameter costs, demonstrating our proposed tuning technique's strong capability and effectiveness in the low-data regime.
Authors: Hongbing Zhang
Abstract: Low-rank tensor completion (LRTC) aims to recover a complete low-rank tensor from incomplete observed tensor, attracting extensive attention in various practical applications such as image processing and computer vision. However, current methods often perform well only when there is a sufficient of observed information, and they perform poorly or may fail when the observed information is less than 5\%. In order to improve the utilization of observed information, a new method called the tensor joint rank with logarithmic composite norm (TJLC) method is proposed. This method simultaneously exploits two types of tensor low-rank structures, namely tensor Tucker rank and tubal rank, thereby enhancing the inherent correlations between known and missing elements. To address the challenge of applying two tensor ranks with significantly different directly to LRTC, a new tensor Logarithmic composite norm is further proposed. Subsequently, the TJLC model and algorithm for the LRTC problem are proposed. Additionally, theoretical convergence guarantees for the TJLC method are provided. Experiments on various real datasets demonstrate that the proposed method outperforms state-of-the-art methods significantly. Particularly, the proposed method achieves satisfactory recovery even when the observed information is as low as 1\%, and the recovery performance improves significantly as the observed information increases.
Authors: Xun Lin, Wenzhong Tang, Haoran Wang, Yizhong Liu, Yakun Ju, Shuai Wang, Zitong Yu
Abstract: Recently, a surge in scientific publications suspected of image manipulation has led to numerous retractions, bringing the issue of image integrity into sharp focus. Although research on forensic detectors for image plagiarism and image synthesis exists, the detection of image splicing traces in scientific publications remains unexplored. Compared to image duplication and synthesis, image splicing detection is more challenging due to the lack of reference images and the typically small tampered areas. Furthermore, disruptive factors in scientific images, such as artifacts from digital compression, abnormal patterns, and noise from physical operations, present misleading features like splicing traces, significantly increasing the difficulty of this task. Moreover, the scarcity of high-quality datasets of spliced scientific images limits potential advancements. In this work, we propose an Uncertainty-guided Refinement Network (URN) to mitigate the impact of these disruptive factors. Our URN can explicitly suppress the propagation of unreliable information flow caused by disruptive factors between regions, thus obtaining robust splicing features. Additionally, the URN is designed to concentrate improvements in uncertain prediction areas during the decoding phase. We also construct a dataset for image splicing detection (SciSp) containing 1,290 spliced images. Compared to existing datasets, SciSp includes the largest number of spliced images and the most diverse sources. Comprehensive experiments conducted on three benchmark datasets demonstrate the superiority of our approach. We also validate the URN's generalisability in resisting cross-dataset domain shifts and its robustness against various post-processing techniques, including advanced deep-learning-based inpainting.
Authors: Qiang Li, Dan Zhang, Shengzhao Lei, Xun Zhao, Porawit Kamnoedboon, WeiWei Li, Junhao Dong, Shuyan Li
Abstract: Despite the promising performance of existing visual models on public benchmarks, the critical assessment of their robustness for real-world applications remains an ongoing challenge. To bridge this gap, we propose an explainable visual dataset, XIMAGENET-12, to evaluate the robustness of visual models. XIMAGENET-12 consists of over 200K images with 15,410 manual semantic annotations. Specifically, we deliberately selected 12 categories from ImageNet, representing objects commonly encountered in practical life. To simulate real-world situations, we incorporated six diverse scenarios, such as overexposure, blurring, and color changes, etc. We further develop a quantitative criterion for robustness assessment, allowing for a nuanced understanding of how visual models perform under varying conditions, notably in relation to the background. We make the XIMAGENET-12 dataset and its corresponding code openly accessible at \url{https://sites.google.com/view/ximagenet-12/home}. We expect the introduction of the XIMAGENET-12 dataset will empower researchers to thoroughly evaluate the robustness of their visual models under challenging conditions.
Authors: Anuja Vats, David V\"olgyes, Martijn Vermeer, Marius Pedersen, Kiran Raja, Daniele S. M. Fantin, Jacob Alexander Hay
Abstract: Estimating building footprint maps from geospatial data is of paramount importance in urban planning, development, disaster management, and various other applications. Deep learning methodologies have gained prominence in building segmentation maps, offering the promise of precise footprint extraction without extensive post-processing. However, these methods face challenges in generalization and label efficiency, particularly in remote sensing, where obtaining accurate labels can be both expensive and time-consuming. To address these challenges, we propose terrain-aware self-supervised learning, tailored to remote sensing, using digital elevation models from LiDAR data. We propose to learn a model to differentiate between bare Earth and superimposed structures enabling the network to implicitly learn domain-relevant features without the need for extensive pixel-level annotations. We test the effectiveness of our approach by evaluating building segmentation performance on test datasets with varying label fractions. Remarkably, with only 1% of the labels (equivalent to 25 labeled examples), our method improves over ImageNet pre-training, showing the advantage of leveraging unlabeled data for feature extraction in the domain of remote sensing. The performance improvement is more pronounced in few-shot scenarios and gradually closes the gap with ImageNet pre-training as the label fraction increases. We test on a dataset characterized by substantial distribution shifts and labeling errors to demonstrate the generalizability of our approach. When compared to other baselines, including ImageNet pretraining and more complex architectures, our approach consistently performs better, demonstrating the efficiency and effectiveness of self-supervised terrain-aware feature learning.
Authors: Matthieu Armando, Salma Galaaoui, Fabien Baradel, Thomas Lucas, Vincent Leroy, Romain Br\'egier, Philippe Weinzaepfel, Gr\'egory Rogez
Abstract: Human perception and understanding is a major domain of computer vision which, like many other vision subdomains recently, stands to gain from the use of large models pre-trained on large datasets. We hypothesize that the most common pre-training strategy of relying on general purpose, object-centric image datasets such as ImageNet, is limited by an important domain shift. On the other hand, collecting domain-specific ground truth such as 2D or 3D labels does not scale well. Therefore, we propose a pre-training approach based on self-supervised learning that works on human-centric data using only images. Our method uses pairs of images of humans: the first is partially masked and the model is trained to reconstruct the masked parts given the visible ones and a second image. It relies on both stereoscopic (cross-view) pairs, and temporal (cross-pose) pairs taken from videos, in order to learn priors about 3D as well as human motion. We pre-train a model for body-centric tasks and one for hand-centric tasks. With a generic transformer architecture, these models outperform existing self-supervised pre-training methods on a wide set of human-centric downstream tasks, and obtain state-of-the-art performance for instance when fine-tuning for model-based and model-free human mesh recovery.
Authors: Adam Tonderski, Carl Lindstr\"om, Georg Hess, William Ljungbergh, Lennart Svensson, Christoffer Petersson
Abstract: Neural radiance fields (NeRFs) have gained popularity in the autonomous driving (AD) community. Recent methods show NeRFs' potential for closed-loop simulation, enabling testing of AD systems, and as an advanced training data augmentation technique. However, existing methods often require long training times, dense semantic supervision, or lack generalizability. This, in turn, hinders the application of NeRFs for AD at scale. In this paper, we propose NeuRAD, a robust novel view synthesis method tailored to dynamic AD data. Our method features simple network design, extensive sensor modeling for both camera and lidar -- including rolling shutter, beam divergence and ray dropping -- and is applicable to multiple datasets out of the box. We verify its performance on five popular AD datasets, achieving state-of-the-art performance across the board. To encourage further development, we will openly release the NeuRAD source code. See https://github.com/georghess/NeuRAD .
Authors: Wooseok Kim, Taiki Fukiage, Takeshi Oishi
Abstract: Recently, significant progress has been made in the study of methods for 3D reconstruction from multiple images using implicit neural representations, exemplified by the neural radiance field (NeRF) method. Such methods, which are based on volume rendering, can model various light phenomena, and various extended methods have been proposed to accommodate different scenes and situations. However, when handling scenes with multiple glass objects, e.g., objects in a glass showcase, modeling the target scene accurately has been challenging due to the presence of multiple reflection and refraction effects. Thus, this paper proposes a NeRF-based modeling method for scenes containing a glass case. In the proposed method, refraction and reflection are modeled using elements that are dependent and independent of the viewer's perspective. This approach allows us to estimate the surfaces where refraction occurs, i.e., glass surfaces, and enables the separation and modeling of both direct and reflected light components. The proposed method requires predetermined camera poses, but accurately estimating these poses in scenes with glass objects is difficult. Therefore, we used a robotic arm with an attached camera to acquire images with known poses. Compared to existing methods, the proposed method enables more accurate modeling of both glass refraction and the overall scene.
Authors: Zhixiang Wei, Lin Chen, Yi Jin, Xiaoxiao Ma, Tianle Liu, Pengyang Ling, Ben Wang, Huaian Chen, Jinjin Zheng
Abstract: In this paper, we first assess and harness various Vision Foundation Models (VFMs) in the context of Domain Generalized Semantic Segmentation (DGSS). Driven by the motivation that Leveraging Stronger pre-trained models and Fewer trainable parameters for Superior generalizability, we introduce a robust fine-tuning approach, namely Rein, to parameter-efficiently harness VFMs for DGSS. Built upon a set of trainable tokens, each linked to distinct instances, Rein precisely refines and forwards the feature maps from each layer to the next layer within the backbone. This process produces diverse refinements for different categories within a single image. With fewer trainable parameters, Rein efficiently fine-tunes VFMs for DGSS tasks, surprisingly surpassing full parameter fine-tuning. Extensive experiments across various settings demonstrate that Rein significantly outperforms state-of-the-art methods. Remarkably, with just an extra 1% of trainable parameters within the frozen backbone, Rein achieves a mIoU of 78.4% on the Cityscapes, without accessing any real urban-scene datasets.Code is available at https://github.com/w1oves/Rein.git.
Authors: Yiduo Hao, Sohrab Madani, Junfeng Guan, Mohammed Alloulah, Saurabh Gupta, Haitham Hassanieh
Abstract: The perception of autonomous vehicles using radars has attracted increased research interest due its ability to operate in fog and bad weather. However, training radar models is hindered by the cost and difficulty of annotating large-scale radar data. To overcome this bottleneck, we propose a self-supervised learning framework to leverage the large amount of unlabeled radar data to pre-train radar-only embeddings for self-driving perception tasks. The proposed method combines radar-to-radar and radar-to-vision contrastive losses to learn a general representation from unlabeled radar heatmaps paired with their corresponding camera images. When used for downstream object detection, we demonstrate that the proposed self-supervision framework can improve the accuracy of state-of-the-art supervised baselines by $5.8\%$ in mAP. Code is available at \url{https://github.com/yiduohao/Radical}.
Authors: Yu Chen, Shuai Zheng, Menglong Jin, Yan Chang, Nianyi Wang
Abstract: Fluid motion can be considered as a point cloud transformation when using the SPH method. Compared to traditional numerical analysis methods, using machine learning techniques to learn physics simulations can achieve near-accurate results, while significantly increasing efficiency. In this paper, we propose an innovative approach for 3D fluid simulations utilizing an Attention-based Dual-pipeline Network, which employs a dual-pipeline architecture, seamlessly integrated with an Attention-based Feature Fusion Module. Unlike previous methods, which often make difficult trade-offs between global fluid control and physical law constraints, we find a way to achieve a better balance between these two crucial aspects with a well-designed dual-pipeline approach. Additionally, we design a Type-aware Input Module to adaptively recognize particles of different types and perform feature fusion afterward, such that fluid-solid coupling issues can be better dealt with. Furthermore, we propose a new dataset, Tank3D, to further explore the network's ability to handle more complicated scenes. The experiments demonstrate that our approach not only attains a quantitative enhancement in various metrics, surpassing the state-of-the-art methods but also signifies a qualitative leap in neural network-based simulation by faithfully adhering to the physical laws. Code and video demonstrations are available at https://github.com/chenyu-xjtu/DualFluidNet.
Authors: Shengyi Qian, Weifeng Chen, Min Bai, Xiong Zhou, Zhuowen Tu, Li Erran Li
Abstract: Affordance grounding refers to the task of finding the area of an object with which one can interact. It is a fundamental but challenging task, as a successful solution requires the comprehensive understanding of a scene in multiple aspects including detection, localization, and recognition of objects with their parts, of geo-spatial configuration/layout of the scene, of 3D shapes and physics, as well as of the functionality and potential interaction of the objects and humans. Much of the knowledge is hidden and beyond the image content with the supervised labels from a limited training set. In this paper, we make an attempt to improve the generalization capability of the current affordance grounding by taking the advantage of the rich world, abstract, and human-object-interaction knowledge from pretrained large-scale vision language models. Under the AGD20K benchmark, our proposed model demonstrates a significant performance gain over the competing methods for in-the-wild object affordance grounding. We further demonstrate it can ground affordance for objects from random Internet images, even if both objects and actions are unseen during training. Project site: https://jasonqsy.github.io/AffordanceLLM/
Authors: Shun Fang, Ming Cui, Xing Feng, Yanna Lv
Abstract: Neural Radiation Field (NeRF) technology can learn a 3D implicit model of a scene from 2D images and synthesize realistic novel view images. This technology has received widespread attention from the industry and has good application prospects. In response to the problem that the rendering quality of NeRF images needs to be improved, many researchers have proposed various methods to improve the rendering quality in the past three years. The latest relevant papers are classified and reviewed, the technical principles behind quality improvement are analyzed, and the future evolution direction of quality improvement methods is discussed. This study can help researchers quickly understand the current state and evolutionary context of technology in this field, which is helpful in inspiring the development of more efficient algorithms and promoting the application of NeRF technology in related fields.
Authors: Yanhua Zhang, Ke Zhang, Jingyu Wang, Yulin Wu, Wuwei Wang
Abstract: Real-time semantic segmentation is a crucial research for real-world applications. However, many methods lay particular emphasis on reducing the computational complexity and model size, while largely sacrificing the accuracy. To tackle this problem, we propose a parallel inference network customized for semantic segmentation tasks to achieve a good trade-off between speed and accuracy. We employ a shallow backbone to ensure real-time speed, and propose three core components to compensate for the reduced model capacity to improve accuracy. Specifically, we first design a dual-pyramidal path architecture (Multi-level Feature Aggregation Module, MFAM) to aggregate multi-level features from the encoder to each scale, providing hierarchical clues for subsequent spatial alignment and corresponding in-network inference. Then, we build Recursive Alignment Module (RAM) by combining the flow-based alignment module with recursive upsampling architecture for accurate spatial alignment between multi-scale feature maps with half the computational complexity of the straightforward alignment method. Finally, we perform independent parallel inference on the aligned features to obtain multi-scale scores, and adaptively fuse them through an attention-based Adaptive Scores Fusion Module (ASFM) so that the final prediction can favor objects of multiple scales. Our framework shows a better balance between speed and accuracy than state-of-the-art real-time methods on Cityscapes and CamVid datasets. We also conducted systematic ablation studies to gain insight into our motivation and architectural design. Code is available at: https://github.com/Yanhua-Zhang/MFARANet.
Authors: Siming Yan, Min Bai, Weifeng Chen, Xiong Zhou, Qixing Huang, Li Erran Li
Abstract: By combining natural language understanding, generation capabilities, and breadth of knowledge of large language models with image perception, recent large vision language models (LVLMs) have shown unprecedented visual reasoning capabilities. However, the generated text often suffers from inaccurate grounding in the visual input, resulting in errors such as hallucination of nonexistent scene elements, missing significant parts of the scene, and inferring incorrect attributes of and relationships between objects. To address these issues, we introduce a novel framework, ViGoR(Visual Grounding Through Fine-Grained Reward Modeling) that utilizes fine-grained reward modeling to significantly enhance the visual grounding of LVLMs over pre-trained baselines. This improvement is efficiently achieved using much cheaper human evaluations instead of full supervisions, as well as automated methods. We show the effectiveness of our approach through a variety of evaluation methods and benchmarks. Additionally, we plan to release our human annotation comprising approximately 16,000 images and generated text pairs with fine-grained evaluations to contribute to related research in the community.
Authors: Zequn Yang, Yake Wei, Ce Liang, Di Hu
Abstract: Multi-modal models have shown a promising capability to effectively integrate information from various sources, yet meanwhile, they are found vulnerable to pervasive perturbations, such as uni-modal attacks and missing conditions. To counter these perturbations, robust multi-modal representations are highly expected, which are positioned well away from the discriminative multi-modal decision boundary. In this paper, different from conventional empirical studies, we focus on a commonly used joint multi-modal framework and theoretically discover that larger uni-modal representation margins and more reliable integration for modalities are essential components for achieving higher robustness. This discovery can further explain the limitation of multi-modal robustness and the phenomenon that multi-modal models are often vulnerable to attacks on the specific modality. Moreover, our analysis reveals how the widespread issue, that the model has different preferences for modalities, limits the multi-modal robustness by influencing the essential components and could lead to attacks on the specific modality highly effective. Inspired by our theoretical finding, we introduce a training procedure called Certifiable Robust Multi-modal Training (CRMT), which can alleviate this influence from modality preference and explicitly regulate essential components to significantly improve robustness in a certifiable manner. Our method demonstrates substantial improvements in performance and robustness compared with existing methods. Furthermore, our training procedure can be easily extended to enhance other robust training strategies, highlighting its credibility and flexibility.
Authors: Nikola Zubi\'c, Mathias Gehrig, Davide Scaramuzza
Abstract: Today, state-of-the-art deep neural networks that process event-camera data first convert a temporal window of events into dense, grid-like input representations. As such, they exhibit poor generalizability when deployed at higher inference frequencies (i.e., smaller temporal windows) than the ones they were trained on. We address this challenge by introducing state-space models (SSMs) with learnable timescale parameters to event-based vision. This design adapts to varying frequencies without the need to retrain the network at different frequencies. Additionally, we investigate two strategies to counteract aliasing effects when deploying the model at higher frequencies. We comprehensively evaluate our approach against existing methods based on RNN and Transformer architectures across various benchmarks, including Gen1 and 1 Mpx event camera datasets. Our results demonstrate that SSM-based models train 33% faster and also exhibit minimal performance degradation when tested at higher frequencies than the training input. Traditional RNN and Transformer models exhibit performance drops of more than 20 mAP, with SSMs having a drop of 3.76 mAP, highlighting the effectiveness of SSMs in event-based vision tasks.
Authors: Yixin Liu, Kai Zhang, Yuan Li, Zhiling Yan, Chujie Gao, Ruoxi Chen, Zhengqing Yuan, Yue Huang, Hanchi Sun, Jianfeng Gao, Lifang He, Lichao Sun
Abstract: Sora is a text-to-video generative AI model, released by OpenAI in February 2024. The model is trained to generate videos of realistic or imaginative scenes from text instructions and show potential in simulating the physical world. Based on public technical reports and reverse engineering, this paper presents a comprehensive review of the model's background, related technologies, applications, remaining challenges, and future directions of text-to-video AI models. We first trace Sora's development and investigate the underlying technologies used to build this "world simulator". Then, we describe in detail the applications and potential impact of Sora in multiple industries ranging from film-making and education to marketing. We discuss the main challenges and limitations that need to be addressed to widely deploy Sora, such as ensuring safe and unbiased video generation. Lastly, we discuss the future development of Sora and video generation models in general, and how advancements in the field could enable new ways of human-AI interaction, boosting productivity and creativity of video generation.
Authors: Francisco Raverta Capua, Juan Schandin, Pablo De Crist\'oforis
Abstract: Remote sensing through unmanned aerial systems (UAS) has been increasing in forestry in recent years, along with using machine learning for data processing. Deep learning architectures, extensively applied in natural language and image processing, have recently been extended to the point cloud domain. However, the availability of point cloud datasets for training and testing remains limited. Creating forested environment point cloud datasets is expensive, requires high-precision sensors, and is time-consuming as manual point classification is required. Moreover, forest areas could be inaccessible or dangerous for humans, further complicating data collection. Then, a question arises whether it is possible to use synthetic data to train deep learning networks without the need to rely on large volumes of real forest data. To answer this question, we developed a realistic simulator that procedurally generates synthetic forest scenes. Thanks to this, we have conducted a comparative study of different state-of-the-art point-based deep learning networks for forest segmentation. Using created datasets, we determined the feasibility of using synthetic data to train deep learning networks to classify point clouds from real forest datasets. Both the simulator and the datasets are released as part of this work.
Authors: Md Mushfiqur Azam, Kevin Desai
Abstract: Egocentric human pose estimation aims to estimate human body poses and develop body representations from a first-person camera perspective. It has gained vast popularity in recent years because of its wide range of applications in sectors like XR-technologies, human-computer interaction, and fitness tracking. However, to the best of our knowledge, there is no systematic literature review based on the proposed solutions regarding egocentric 3D human pose estimation. To that end, the aim of this survey paper is to provide an extensive overview of the current state of egocentric pose estimation research. In this paper, we categorize and discuss the popular datasets and the different pose estimation models, highlighting the strengths and weaknesses of different methods by comparative analysis. This survey can be a valuable resource for both researchers and practitioners in the field, offering insights into key concepts and cutting-edge solutions in egocentric pose estimation, its wide-ranging applications, as well as the open problems with future scope.
Authors: Qiyuan He, Jinghao Wang, Ziwei Liu, Angela Yao
Abstract: Conditional diffusion models can create unseen images in various settings, aiding image interpolation. Interpolation in latent spaces is well-studied, but interpolation with specific conditions like text or poses is less understood. Simple approaches, such as linear interpolation in the space of conditions, often result in images that lack consistency, smoothness, and fidelity. To that end, we introduce a novel training-free technique named Attention Interpolation via Diffusion (AID). Our key contributions include 1) proposing an inner/outer interpolated attention layer; 2) fusing the interpolated attention with self-attention to boost fidelity; and 3) applying beta distribution to selection to increase smoothness. We also present a variant, Prompt-guided Attention Interpolation via Diffusion (PAID), that considers interpolation as a condition-dependent generative process. This method enables the creation of new images with greater consistency, smoothness, and efficiency, and offers control over the exact path of interpolation. Our approach demonstrates effectiveness for conceptual and spatial interpolation. Code and demo are available at https://github.com/QY-H00/attention-interpolation-diffusion.
URLs: https://github.com/QY-H00/attention-interpolation-diffusion.
Authors: Muhammad Zubair Irshad, Sergey Zakahrov, Vitor Guizilini, Adrien Gaidon, Zsolt Kira, Rares Ambrus
Abstract: Neural fields excel in computer vision and robotics due to their ability to understand the 3D visual world such as inferring semantics, geometry, and dynamics. Given the capabilities of neural fields in densely representing a 3D scene from 2D images, we ask the question: Can we scale their self-supervised pretraining, specifically using masked autoencoders, to generate effective 3D representations from posed RGB images. Owing to the astounding success of extending transformers to novel data modalities, we employ standard 3D Vision Transformers to suit the unique formulation of NeRFs. We leverage NeRF's volumetric grid as a dense input to the transformer, contrasting it with other 3D representations such as pointclouds where the information density can be uneven, and the representation is irregular. Due to the difficulty of applying masked autoencoders to an implicit representation, such as NeRF, we opt for extracting an explicit representation that canonicalizes scenes across domains by employing the camera trajectory for sampling. Our goal is made possible by masking random patches from NeRF's radiance and density grid and employing a standard 3D Swin Transformer to reconstruct the masked patches. In doing so, the model can learn the semantic and spatial structure of complete scenes. We pretrain this representation at scale on our proposed curated posed-RGB data, totaling over 1.6 million images. Once pretrained, the encoder is used for effective 3D transfer learning. Our novel self-supervised pretraining for NeRFs, NeRF-MAE, scales remarkably well and improves performance on various challenging 3D tasks. Utilizing unlabeled posed 2D data for pretraining, NeRF-MAE significantly outperforms self-supervised 3D pretraining and NeRF scene understanding baselines on Front3D and ScanNet datasets with an absolute performance improvement of over 20% AP50 and 8% AP25 for 3D object detection.
Authors: Giuseppe Macario
Abstract: This work proposes a WebXR-based cross-platform conceptual architecture, leveraging the A-Frame and Networked-Aframe frameworks, in order to facilitate the development of an open, accessible, and interoperable metaverse. By introducing the concept of spatial web app, this research contributes to the discourse on the metaverse, offering an architecture that democratizes access to virtual environments and extended reality through the web, and aligns with Tim Berners-Lee's original vision of the World Wide Web as an open platform in the digital realm.
Authors: Xiaoyi Bao, Siyang Sun, Shuailei Ma, Kecheng Zheng, Yuxin Guo, Guosheng Zhao, Yun Zheng, Xingang Wang
Abstract: The reasoning segmentation task, which demands a nuanced comprehension of intricate queries to accurately pinpoint object regions, is attracting increasing attention. However, Multi-modal Large Language Models (MLLM) often find it difficult to accurately localize the objects described in complex reasoning contexts. We believe that the act of reasoning segmentation should mirror the cognitive stages of human visual search, where each step is a progressive refinement of thought toward the final object. Thus we introduce the Chains of Reasoning and Segmenting (CoReS) and find this top-down visual hierarchy indeed enhances the visual search process. Specifically, we propose a dual-chain structure that generates multi-modal, chain-like outputs to aid the segmentation process. Furthermore, to steer the MLLM's outputs into this intended hierarchy, we incorporate in-context inputs as guidance. Extensive experiments demonstrate the superior performance of our CoReS, which surpasses the state-of-the-art method by 7.1\% on the ReasonSeg dataset. Project: https://chain-of-reasoning-and-segmentation.github.io/.
URLs: https://chain-of-reasoning-and-segmentation.github.io/.
Authors: Haocheng Yuan, Ajian Liu, Junze Zheng, Jun Wan, Jiankang Deng, Sergio Escalera, Hugo Jair Escalante, Isabelle Guyon, Zhen Lei
Abstract: Face Anti-Spoofing (FAS) is crucial to safeguard Face Recognition (FR) Systems. In real-world scenarios, FRs are confronted with both physical and digital attacks. However, existing algorithms often address only one type of attack at a time, which poses significant limitations in real-world scenarios where FR systems face hybrid physical-digital threats. To facilitate the research of Unified Attack Detection (UAD) algorithms, a large-scale UniAttackData dataset has been collected. UniAttackData is the largest public dataset for Unified Attack Detection, with a total of 28,706 videos, where each unique identity encompasses all advanced attack types. Based on this dataset, we organized a Unified Physical-Digital Face Attack Detection Challenge to boost the research in Unified Attack Detections. It attracted 136 teams for the development phase, with 13 qualifying for the final round. The results re-verified by the organizing team were used for the final ranking. This paper comprehensively reviews the challenge, detailing the dataset introduction, protocol definition, evaluation criteria, and a summary of published results. Finally, we focus on the detailed analysis of the highest-performing algorithms and offer potential directions for unified physical-digital attack detection inspired by this competition. Challenge Website: https://sites.google.com/view/face-anti-spoofing-challenge/welcome/challengecvpr2024.
URLs: https://sites.google.com/view/face-anti-spoofing-challenge/welcome/challengecvpr2024.
Authors: Yujie Li, Yanbin Wang, Haitao Xu, Bin Liu, Jianguo Sun, Zhenhao Guo, Wenrui Ma
Abstract: Adversarial attacks induce misclassification by introducing subtle perturbations. Recently, diffusion models are applied to the image classifiers to improve adversarial robustness through adversarial training or by purifying adversarial noise. However, diffusion-based adversarial training often encounters convergence challenges and high computational expenses. Additionally, diffusion-based purification inevitably causes data shift and is deemed susceptible to stronger adaptive attacks. To tackle these issues, we propose the Truth Maximization Diffusion Classifier (TMDC), a generative Bayesian classifier that builds upon pre-trained diffusion models and the Bayesian theorem. Unlike data-driven classifiers, TMDC, guided by Bayesian principles, utilizes the conditional likelihood from diffusion models to determine the class probabilities of input images, thereby insulating against the influences of data shift and the limitations of adversarial training. Moreover, to enhance TMDC's resilience against more potent adversarial attacks, we propose an optimization strategy for diffusion classifiers. This strategy involves post-training the diffusion model on perturbed datasets with ground-truth labels as conditions, guiding the diffusion model to learn the data distribution and maximizing the likelihood under the ground-truth labels. The proposed method achieves state-of-the-art performance on the CIFAR10 dataset against heavy white-box attacks and strong adaptive attacks. Specifically, TMDC achieves robust accuracies of 82.81% against $l_{\infty}$ norm-bounded perturbations and 86.05% against $l_{2}$ norm-bounded perturbations, respectively, with $\epsilon=0.05$.
Authors: Shubham Tiwari, Yash Sethia, Ritesh Kumar, Ashwani Tanwar, Rudresh Dwivedi
Abstract: With the advent of social media, fun selfie filters have come into tremendous mainstream use affecting the functioning of facial biometric systems as well as image recognition systems. These filters vary from beautification filters and Augmented Reality (AR)-based filters to filters that modify facial landmarks. Hence, there is a need to assess the impact of such filters on the performance of existing face recognition systems. The limitation associated with existing solutions is that these solutions focus more on the beautification filters. However, the current AR-based filters and filters which distort facial key points are in vogue recently and make the faces highly unrecognizable even to the naked eye. Also, the filters considered are mostly obsolete with limited variations. To mitigate these limitations, we aim to perform a holistic impact analysis of the latest filters and propose an user recognition model with the filtered images. We have utilized a benchmark dataset for baseline images, and applied the latest filters over them to generate a beautified/filtered dataset. Next, we have introduced a model FaceFilterNet for beautified user recognition. In this framework, we also utilize our model to comment on various attributes of the person including age, gender, and ethnicity. In addition, we have also presented a filter-wise impact analysis on face recognition, age estimation, gender, and ethnicity prediction. The proposed method affirms the efficacy of our dataset with an accuracy of 87.25% and an optimal accuracy for facial attribute analysis.
Authors: Yipo Huang, Xiangfei Sheng, Zhichao Yang, Quan Yuan, Zhichao Duan, Pengfei Chen, Leida Li, Weisi Lin, Guangming Shi
Abstract: The highly abstract nature of image aesthetics perception (IAP) poses significant challenge for current multimodal large language models (MLLMs). The lack of human-annotated multi-modality aesthetic data further exacerbates this dilemma, resulting in MLLMs falling short of aesthetics perception capabilities. To address the above challenge, we first introduce a comprehensively annotated Aesthetic Multi-Modality Instruction Tuning (AesMMIT) dataset, which serves as the footstone for building multi-modality aesthetics foundation models. Specifically, to align MLLMs with human aesthetics perception, we construct a corpus-rich aesthetic critique database with 21,904 diverse-sourced images and 88K human natural language feedbacks, which are collected via progressive questions, ranging from coarse-grained aesthetic grades to fine-grained aesthetic descriptions. To ensure that MLLMs can handle diverse queries, we further prompt GPT to refine the aesthetic critiques and assemble the large-scale aesthetic instruction tuning dataset, i.e. AesMMIT, which consists of 409K multi-typed instructions to activate stronger aesthetic capabilities. Based on the AesMMIT database, we fine-tune the open-sourced general foundation models, achieving multi-modality Aesthetic Expert models, dubbed AesExpert. Extensive experiments demonstrate that the proposed AesExpert models deliver significantly better aesthetic perception performances than the state-of-the-art MLLMs, including the most advanced GPT-4V and Gemini-Pro-Vision. Source data will be available at https://github.com/yipoh/AesExpert.
Authors: Qi Guo, Shanmin Pang, Xiaojun Jia, Qing Guo
Abstract: Targeted transfer-based attacks involving adversarial examples pose a significant threat to large visual-language models (VLMs). However, the state-of-the-art (SOTA) transfer-based attacks incur high costs due to excessive iteration counts. Furthermore, the generated adversarial examples exhibit pronounced adversarial noise and demonstrate limited efficacy in evading defense methods such as DiffPure. To address these issues, inspired by score matching, we introduce AdvDiffVLM, which utilizes diffusion models to generate natural, unrestricted adversarial examples. Specifically, AdvDiffVLM employs Adaptive Ensemble Gradient Estimation to modify the score during the diffusion model's reverse generation process, ensuring the adversarial examples produced contain natural adversarial semantics and thus possess enhanced transferability. Simultaneously, to enhance the quality of adversarial examples further, we employ the GradCAM-guided Mask method to disperse adversarial semantics throughout the image, rather than concentrating them in a specific area. Experimental results demonstrate that our method achieves a speedup ranging from 10X to 30X compared to existing transfer-based attack methods, while maintaining superior quality of adversarial examples. Additionally, the generated adversarial examples possess strong transferability and exhibit increased robustness against adversarial defense methods. Notably, AdvDiffVLM can successfully attack commercial VLMs, including GPT-4V, in a black-box manner.
Authors: Chengyang Yan, Donald G. Dansereau
Abstract: The performance of robots in their applications heavily depends on the quality of sensory input. However, designing sensor payloads and their parameters for specific robotic tasks is an expensive process that requires well-established sensor knowledge and extensive experiments with physical hardware. With cameras playing a pivotal role in robotic perception, we introduce a novel end-to-end optimization approach for co-designing a camera with specific robotic tasks by combining derivative-free and gradient-based optimizers. The proposed method leverages recent computer graphics techniques and physical camera characteristics to prototype the camera in software, simulate operational environments and tasks for robots, and optimize the camera design based on the desired tasks in a cost-effective way. We validate the accuracy of our camera simulation by comparing it with physical cameras, and demonstrate the design of cameras with stronger performance than common off-the-shelf alternatives. Our approach supports the optimization of both continuous and discrete camera parameters, manufacturing constraints, and can be generalized to a broad range of camera design scenarios including multiple cameras and unconventional cameras. This work advances the fully automated design of cameras for specific robotics tasks.
Authors: Mohammad Shiri, Monalika Padma Reddy, Jiangwen Sun
Abstract: Invasive ductal carcinoma (IDC) is the most prevalent form of breast cancer. Breast tissue histopathological examination is critical in diagnosing and classifying breast cancer. Although existing methods have shown promising results, there is still room for improvement in the classification accuracy and generalization of IDC using histopathology images. We present a novel approach, Supervised Contrastive Vision Transformer (SupCon-ViT), for improving the classification of invasive ductal carcinoma in terms of accuracy and generalization by leveraging the inherent strengths and advantages of both transfer learning, i.e., pre-trained vision transformer, and supervised contrastive learning. Our results on a benchmark breast cancer dataset demonstrate that SupCon-Vit achieves state-of-the-art performance in IDC classification, with an F1-score of 0.8188, precision of 0.7692, and specificity of 0.8971, outperforming existing methods. In addition, the proposed model demonstrates resilience in scenarios with minimal labeled data, making it highly efficient in real-world clinical settings where labelled data is limited. Our findings suggest that supervised contrastive learning in conjunction with pre-trained vision transformers appears to be a viable strategy for an accurate classification of IDC, thus paving the way for a more efficient and reliable diagnosis of breast cancer through histopathological image analysis.
Authors: Dingkun Zhang, Sijia Li, Chen Chen, Qingsong Xie, Haonan Lu
Abstract: In the era of AIGC, the demand for low-budget or even on-device applications of diffusion models emerged. In terms of compressing the Stable Diffusion models (SDMs), several approaches have been proposed, and most of them leveraged the handcrafted layer removal methods to obtain smaller U-Nets, along with knowledge distillation to recover the network performance. However, such a handcrafting manner of layer removal is inefficient and lacks scalability and generalization, and the feature distillation employed in the retraining phase faces an imbalance issue that a few numerically significant feature loss terms dominate over others throughout the retraining process. To this end, we proposed the layer pruning and normalized distillation for compressing diffusion models (LAPTOP-Diff). We, 1) introduced the layer pruning method to compress SDM's U-Net automatically and proposed an effective one-shot pruning criterion whose one-shot performance is guaranteed by its good additivity property, surpassing other layer pruning and handcrafted layer removal methods, 2) proposed the normalized feature distillation for retraining, alleviated the imbalance issue. Using the proposed LAPTOP-Diff, we compressed the U-Nets of SDXL and SDM-v1.5 for the most advanced performance, achieving a minimal 4.0% decline in PickScore at a pruning ratio of 50% while the comparative methods' minimal PickScore decline is 8.2%. We will release our code.
Authors: Qiangang Du, Jinlong Peng, Xu Chen, Qingdong He, Liren He, Qiang Nie, Wenbing Zhu, Mingmin Chi, Yabiao Wang, Chengjie Wang
Abstract: Change detection is widely applied in remote sensing image analysis. Existing methods require training models separately for each dataset, which leads to poor domain generalization. Moreover, these methods rely heavily on large amounts of high-quality pair-labelled data for training, which is expensive and impractical. In this paper, we propose a multimodal contrastive learning (ChangeCLIP) based on visual-language pre-training for change detection domain generalization. Additionally, we propose a dynamic context optimization for prompt learning. Meanwhile, to address the data dependency issue of existing methods, we introduce a single-temporal and controllable AI-generated training strategy (SAIN). This allows us to train the model using a large number of single-temporal images without image pairs in the real world, achieving excellent generalization. Extensive experiments on series of real change detection datasets validate the superiority and strong generalization of ChangeCLIP, outperforming state-of-the-art change detection methods. Code will be available.
Authors: Jeongtaek Oh, Jaeyoung Chung, Dongwoo Lee, Kyoung Mu Lee
Abstract: Although significant progress has been made in reconstructing sharp 3D scenes from motion-blurred images, a transition to real-world applications remains challenging. The primary obstacle stems from the severe blur which leads to inaccuracies in the acquisition of initial camera poses through Structure-from-Motion, a critical aspect often overlooked by previous approaches. To address this challenge, we propose DeblurGS, a method to optimize sharp 3D Gaussian Splatting from motion-blurred images, even with the noisy camera pose initialization. We restore a fine-grained sharp scene by leveraging the remarkable reconstruction capability of 3D Gaussian Splatting. Our approach estimates the 6-Degree-of-Freedom camera motion for each blurry observation and synthesizes corresponding blurry renderings for the optimization process. Furthermore, we propose Gaussian Densification Annealing strategy to prevent the generation of inaccurate Gaussians at erroneous locations during the early training stages when camera motion is still imprecise. Comprehensive experiments demonstrate that our DeblurGS achieves state-of-the-art performance in deblurring and novel view synthesis for real-world and synthetic benchmark datasets, as well as field-captured blurry smartphone videos.
Authors: Zeyu Zhang, Xuyin Qi, Mingxi Chen, Guangxi Li, Ryan Pham, Ayub Qassim, Ella Berry, Zhibin Liao, Owen Siggs, Robert Mclaughlin, Jamie Craig, Minh-Son To
Abstract: The oxygen saturation level in the blood (SaO2) is crucial for health, particularly in relation to sleep-related breathing disorders. However, continuous monitoring of SaO2 is time-consuming and highly variable depending on patients' conditions. Recently, optical coherence tomography angiography (OCTA) has shown promising development in rapidly and effectively screening eye-related lesions, offering the potential for diagnosing sleep-related disorders. To bridge this gap, our paper presents three key contributions. Firstly, we propose JointViT, a novel model based on the Vision Transformer architecture, incorporating a joint loss function for supervision. Secondly, we introduce a balancing augmentation technique during data preprocessing to improve the model's performance, particularly on the long-tail distribution within the OCTA dataset. Lastly, through comprehensive experiments on the OCTA dataset, our proposed method significantly outperforms other state-of-the-art methods, achieving improvements of up to 12.28% in overall accuracy. This advancement lays the groundwork for the future utilization of OCTA in diagnosing sleep-related disorders. See project website https://steve-zeyu-zhang.github.io/JointViT
Authors: Zichen Liu, Yihao Meng, Hao Ouyang, Yue Yu, Bolin Zhao, Daniel Cohen-Or, Huamin Qu
Abstract: Text animation serves as an expressive medium, transforming static communication into dynamic experiences by infusing words with motion to evoke emotions, emphasize meanings, and construct compelling narratives. Crafting animations that are semantically aware poses significant challenges, demanding expertise in graphic design and animation. We present an automated text animation scheme, termed "Dynamic Typography", which combines two challenging tasks. It deforms letters to convey semantic meaning and infuses them with vibrant movements based on user prompts. Our technique harnesses vector graphics representations and an end-to-end optimization-based framework. This framework employs neural displacement fields to convert letters into base shapes and applies per-frame motion, encouraging coherence with the intended textual concept. Shape preservation techniques and perceptual loss regularization are employed to maintain legibility and structural integrity throughout the animation process. We demonstrate the generalizability of our approach across various text-to-video models and highlight the superiority of our end-to-end methodology over baseline methods, which might comprise separate tasks. Through quantitative and qualitative evaluations, we demonstrate the effectiveness of our framework in generating coherent text animations that faithfully interpret user prompts while maintaining readability. Our code is available at: https://animate-your-word.github.io/demo/.
Authors: Haoyue Zhang, Jennifer S. Polson, Eric J. Yang, Kambiz Nael, William Speier, Corey W. Arnold
Abstract: For acute ischemic stroke (AIS) patients with large vessel occlusions, clinicians must decide if the benefit of mechanical thrombectomy (MTB) outweighs the risks and potential complications following an invasive procedure. Pre-treatment computed tomography (CT) and angiography (CTA) are widely used to characterize occlusions in the brain vasculature. If a patient is deemed eligible, a modified treatment in cerebral ischemia (mTICI) score will be used to grade how well blood flow is reestablished throughout and following the MTB procedure. An estimation of the likelihood of successful recanalization can support treatment decision-making. In this study, we proposed a fully automated prediction of a patient's recanalization score using pre-treatment CT and CTA imaging. We designed a spatial cross attention network (SCANet) that utilizes vision transformers to localize to pertinent slices and brain regions. Our top model achieved an average cross-validated ROC-AUC of 77.33 $\pm$ 3.9\%. This is a promising result that supports future applications of deep learning on CT and CTA for the identification of eligible AIS patients for MTB.
Authors: James Gornet, Matthew Thomson
Abstract: Humans construct internal cognitive maps of their environment directly from sensory inputs without access to a system of explicit coordinates or distance measurements. While machine learning algorithms like SLAM utilize specialized visual inference procedures to identify visual features and construct spatial maps from visual and odometry data, the general nature of cognitive maps in the brain suggests a unified mapping algorithmic strategy that can generalize to auditory, tactile, and linguistic inputs. Here, we demonstrate that predictive coding provides a natural and versatile neural network algorithm for constructing spatial maps using sensory data. We introduce a framework in which an agent navigates a virtual environment while engaging in visual predictive coding using a self-attention-equipped convolutional neural network. While learning a next image prediction task, the agent automatically constructs an internal representation of the environment that quantitatively reflects distances. The internal map enables the agent to pinpoint its location relative to landmarks using only visual information.The predictive coding network generates a vectorized encoding of the environment that supports vector navigation where individual latent space units delineate localized, overlapping neighborhoods in the environment. Broadly, our work introduces predictive coding as a unified algorithmic framework for constructing cognitive maps that can naturally extend to the mapping of auditory, sensorimotor, and linguistic inputs.
Authors: Utkarsh Oggy Sarawgi, John Berkowitz, Vineet Garg, Arnav Kundu, Minsik Cho, Sai Srujana Buddi, Saurabh Adya, Ahmed Tewfik
Abstract: Streaming neural network models for fast frame-wise responses to various speech and sensory signals are widely adopted on resource-constrained platforms. Hence, increasing the learning capacity of such streaming models (i.e., by adding more parameters) to improve the predictive power may not be viable for real-world tasks. In this work, we propose a new loss, Streaming Anchor Loss (SAL), to better utilize the given learning capacity by encouraging the model to learn more from essential frames. More specifically, our SAL and its focal variations dynamically modulate the frame-wise cross entropy loss based on the importance of the corresponding frames so that a higher loss penalty is assigned for frames within the temporal proximity of semantically critical events. Therefore, our loss ensures that the model training focuses on predicting the relatively rare but task-relevant frames. Experimental results with standard lightweight convolutional and recurrent streaming networks on three different speech based detection tasks demonstrate that SAL enables the model to learn the overall task more effectively with improved accuracy and latency, without any additional data, model parameters, or architectural changes.
Authors: Siyuan Cheng, Bozhong Tian, Qingbin Liu, Xi Chen, Yongheng Wang, Huajun Chen, Ningyu Zhang
Abstract: In this paper, we focus on editing Multimodal Large Language Models (MLLMs). Compared to editing single-modal LLMs, multimodal model editing is more challenging, which demands a higher level of scrutiny and careful consideration in the editing process. To facilitate research in this area, we construct a new benchmark, dubbed MMEdit, for editing multimodal LLMs and establishing a suite of innovative metrics for evaluation. We conduct comprehensive experiments involving various model editing baselines and analyze the impact of editing different components for multimodal LLMs. Empirically, we notice that previous baselines can implement editing multimodal LLMs to some extent, but the effect is still barely satisfactory, indicating the potential difficulty of this task. We hope that our work can provide the NLP community with insights. Code and dataset are available in https://github.com/zjunlp/EasyEdit.
Authors: Deborah Pereg
Abstract: We introduce Back to Basics (BTB), a fast iterative algorithm for noise reduction. Our method is computationally efficient, does not require training or ground truth data, and can be applied in the presence of independent noise, as well as correlated (coherent) noise, where the noise level is unknown. We examine three study cases: natural image denoising in the presence of additive white Gaussian noise, Poisson-distributed image denoising, and speckle suppression in optical coherence tomography (OCT). Experimental results demonstrate that the proposed approach can effectively improve image quality, in challenging noise settings. Theoretical guarantees are provided for convergence stability.
Authors: Yuxuan Shi, Jun Xu, Dinggang Shen
Abstract: Cone Beam Computed Tomography (CBCT) plays a key role in dental diagnosis and surgery. However, the metal teeth implants could bring annoying metal artifacts during the CBCT imaging process, interfering diagnosis and downstream processing such as tooth segmentation. In this paper, we develop an efficient Transformer to perform metal artifacts reduction (MAR) from dental CBCT images. The proposed MAR Transformer (MARformer) reduces computation complexity in the multihead self-attention by a new Dimension-Reduced Self-Attention (DRSA) module, based on that the CBCT images have globally similar structure. A Patch-wise Perceptive Feed Forward Network (P2FFN) is also proposed to perceive local image information for fine-grained restoration. Experimental results on CBCT images with synthetic and real-world metal artifacts show that our MARformer is efficient and outperforms previous MAR methods and two restoration Transformers.
Authors: Kyra Ahrens, Hans Hergen Lehmann, Jae Hee Lee, Stefan Wermter
Abstract: We address the Continual Learning (CL) problem, wherein a model must learn a sequence of tasks from non-stationary distributions while preserving prior knowledge upon encountering new experiences. With the advancement of foundation models, CL research has pivoted from the initial learning-from-scratch paradigm towards utilizing generic features from large-scale pre-training. However, existing approaches to CL with pre-trained models primarily focus on separating class-specific features from the final representation layer and neglect the potential of intermediate representations to capture low- and mid-level features, which are more invariant to domain shifts. In this work, we propose LayUP, a new prototype-based approach to continual learning that leverages second-order feature statistics from multiple intermediate layers of a pre-trained network. Our method is conceptually simple, does not require access to prior data, and works out of the box with any foundation model. LayUP surpasses the state of the art in four of the seven class-incremental learning benchmarks, all three domain-incremental learning benchmarks and in six of the seven online continual learning benchmarks, while significantly reducing memory and computational requirements compared to existing baselines. Our results demonstrate that fully exhausting the representational capacities of pre-trained models in CL goes well beyond their final embeddings.
Authors: Yijie Yang, Qifeng Gao, Yuping Duan
Abstract: The unrolling method has been investigated for learning variational models in X-ray computed tomography. However, it has been observed that directly unrolling the regularization model through gradient descent does not produce satisfactory results. In this paper, we present a novel deep learning-based CT reconstruction model, where the low-resolution image is introduced to obtain an effective regularization term for improving the network`s robustness. Our approach involves constructing the backbone network architecture by algorithm unrolling that is realized using the deep equilibrium architecture. We theoretically discuss the convergence of the proposed low-resolution prior equilibrium model and provide the conditions to guarantee convergence. Experimental results on both sparse-view and limited-angle reconstruction problems are provided, demonstrating that our end-to-end low-resolution prior equilibrium model outperforms other state-of-the-art methods in terms of noise reduction, contrast-to-noise ratio, and preservation of edge details.
Authors: Junyang Wang, Haiyang Xu, Jiabo Ye, Ming Yan, Weizhou Shen, Ji Zhang, Fei Huang, Jitao Sang
Abstract: Mobile device agent based on Multimodal Large Language Models (MLLM) is becoming a popular application. In this paper, we introduce Mobile-Agent, an autonomous multi-modal mobile device agent. Mobile-Agent first leverages visual perception tools to accurately identify and locate both the visual and textual elements within the app's front-end interface. Based on the perceived vision context, it then autonomously plans and decomposes the complex operation task, and navigates the mobile Apps through operations step by step. Different from previous solutions that rely on XML files of Apps or mobile system metadata, Mobile-Agent allows for greater adaptability across diverse mobile operating environments in a vision-centric way, thereby eliminating the necessity for system-specific customizations. To assess the performance of Mobile-Agent, we introduced Mobile-Eval, a benchmark for evaluating mobile device operations. Based on Mobile-Eval, we conducted a comprehensive evaluation of Mobile-Agent. The experimental results indicate that Mobile-Agent achieved remarkable accuracy and completion rates. Even with challenging instructions, such as multi-app operations, Mobile-Agent can still complete the requirements. Code and model will be open-sourced at https://github.com/X-PLUG/MobileAgent.
Authors: Mengdan Zhu, Zhenke Liu, Bo Pan, Abhinav Angirekula, Liang Zhao
Abstract: Learning interpretable representations of data generative latent factors is an important topic for the development of artificial intelligence. With the rise of the large multimodal model, it can align images with text to generate answers. In this work, we propose a framework to comprehensively explain each latent variable in the generative models using a large multimodal model. We further measure the uncertainty of our generated explanations, quantitatively evaluate the performance of explanation generation among multiple large multimodal models, and qualitatively visualize the variations of each latent variable to learn the disentanglement effects of different generative models on explanations. Finally, we discuss the explanatory capabilities and limitations of state-of-the-art large multimodal models.
Authors: Gijs Bellaard, Sei Sakata, Bart M. N. Smets, Remco Duits
Abstract: PDE-based Group Convolutional Neural Networks (PDE-G-CNNs) utilize solvers of geometrically meaningful evolution PDEs as substitutes for the conventional components in G-CNNs. PDE-G-CNNs offer several key benefits all at once: fewer parameters, inherent equivariance, better performance, data efficiency, and geometric interpretability. In this article we focus on Euclidean equivariant PDE-G-CNNs where the feature maps are two dimensional throughout. We call this variant of the framework a PDE-CNN. From a machine learning perspective, we list several practically desirable axioms and derive from these which PDEs should be used in a PDE-CNN. Here our approach to geometric learning via PDEs is inspired by the axioms of classical linear and morphological scale-space theory, which we generalize by introducing semifield-valued signals. Furthermore, we experimentally confirm for small networks that PDE-CNNs offer fewer parameters, increased performance, and better data efficiency when compared to CNNs. We also investigate what effect the use of different semifields has on the performance of the models.
Authors: Ye Wang, Yaxiong Wang, Yujiao Wu, Bingchen Zhao, Xueming Qian
Abstract: Generalized Class Discovery (GCD) aims to dynamically assign labels to unlabelled data partially based on knowledge learned from labelled data, where the unlabelled data may come from known or novel classes. The prevailing approach generally involves clustering across all data and learning conceptions by prototypical contrastive learning. However, existing methods largely hinge on the performance of clustering algorithms and are thus subject to their inherent limitations. Firstly, the estimated cluster number is often smaller than the ground truth, making the existing methods suffer from the lack of prototypes for comprehensive conception learning. To address this issue, we propose an adaptive probing mechanism that introduces learnable potential prototypes to expand cluster prototypes (centers). As there is no ground truth for the potential prototype, we develop a self-supervised prototype learning framework to optimize the potential prototype in an end-to-end fashion. Secondly, clustering is computationally intensive, and the conventional strategy of clustering both labelled and unlabelled instances exacerbates this issue. To counteract this inefficiency, we opt to cluster only the unlabelled instances and subsequently expand the cluster prototypes with our introduced potential prototypes to fast explore novel classes. Despite the simplicity of our proposed method, extensive empirical analysis on a wide range of datasets confirms that our method consistently delivers state-of-the-art results. Specifically, our method surpasses the nearest competitor by a significant margin of \textbf{9.7}$\%$ within the Stanford Cars dataset and \textbf{12$\times$} clustering efficiency within the Herbarium 19 dataset. We will make the code and checkpoints publicly available at \url{https://github.com/xjtuYW/PNP.git}.
Authors: Tobias Weber, Jakob Dexl, David R\"ugamer, Michael Ingrisch
Abstract: We address the computational barrier of deploying advanced deep learning segmentation models in clinical settings by studying the efficacy of network compression through tensor decomposition. We propose a post-training Tucker factorization that enables the decomposition of pre-existing models to reduce computational requirements without impeding segmentation accuracy. We applied Tucker decomposition to the convolutional kernels of the TotalSegmentator (TS) model, an nnU-Net model trained on a comprehensive dataset for automatic segmentation of 117 anatomical structures. Our approach reduced the floating-point operations (FLOPs) and memory required during inference, offering an adjustable trade-off between computational efficiency and segmentation quality. This study utilized the publicly available TS dataset, employing various downsampling factors to explore the relationship between model size, inference speed, and segmentation performance. The application of Tucker decomposition to the TS model substantially reduced the model parameters and FLOPs across various compression rates, with limited loss in segmentation accuracy. We removed up to 88% of the model's parameters with no significant performance changes in the majority of classes after fine-tuning. Practical benefits varied across different graphics processing unit (GPU) architectures, with more distinct speed-ups on less powerful hardware. Post-hoc network compression via Tucker decomposition presents a viable strategy for reducing the computational demand of medical image segmentation models without substantially sacrificing accuracy. This approach enables the broader adoption of advanced deep learning technologies in clinical practice, offering a way to navigate the constraints of hardware capabilities.
Authors: Wei Chen, Zhiyuan Li
Abstract: A multimodal AI agent is characterized by its ability to process and learn from various types of data, including natural language, visual, and audio inputs, to inform its actions. Despite advancements in large language models that incorporate visual data, such as GPT-4V, effectively translating image-based data into actionable outcomes for AI agents continues to be challenging. In this paper, we introduce a multimodal model that incorporates the concept of functional token specifically designed for AI agent applications. To ensure compatibility with edge devices, our model is optimized to a compact size of less than 1B parameters. Like GPT-4, our model can process both English and Chinese. We demonstrate that this model is capable of operating efficiently on a wide range of edge devices, including as constrained as a Raspberry Pi.