Authors: Changqi Sun, Hao Xu, Yuntian Chen, Dongxiao Zhang
Explainable artificial intelligence (XAI) aims to develop transparent explanatory approaches for "black-box" deep learning models. However,it remains difficult for existing methods to achieve the trade-off of the three key criteria in interpretability, namely, reliability, causality, and usability, which hinder their practical applications. In this paper, we propose a self-supervised automatic semantic interpretable explainable artificial intelligence (AS-XAI) framework, which utilizes transparent orthogonal embedding semantic extraction spaces and row-centered principal component analysis (PCA) for global semantic interpretation of model decisions in the absence of human interference, without additional computational costs. In addition, the invariance of filter feature high-rank decomposition is used to evaluate model sensitivity to different semantic concepts. Extensive experiments demonstrate that robust and orthogonal semantic spaces can be automatically extracted by AS-XAI, providing more effective global interpretability for convolutional neural networks (CNNs) and generating human-comprehensible explanations. The proposed approach offers broad fine-grained extensible practical applications, including shared semantic interpretation under out-of-distribution (OOD) categories, auxiliary explanations for species that are challenging to distinguish, and classification explanations from various perspectives.
Authors: Yi-Hua Huang, Yang-Tian Sun, Ziyi Yang, Xiaoyang Lyu, Yan-Pei Cao, Xiaojuan Qi
Novel view synthesis for dynamic scenes is still a challenging problem in computer vision and graphics. Recently, Gaussian splatting has emerged as a robust technique to represent static scenes and enable high-quality and real-time novel view synthesis. Building upon this technique, we propose a new representation that explicitly decomposes the motion and appearance of dynamic scenes into sparse control points and dense Gaussians, respectively. Our key idea is to use sparse control points, significantly fewer in number than the Gaussians, to learn compact 6 DoF transformation bases, which can be locally interpolated through learned interpolation weights to yield the motion field of 3D Gaussians. We employ a deformation MLP to predict time-varying 6 DoF transformations for each control point, which reduces learning complexities, enhances learning abilities, and facilitates obtaining temporal and spatial coherent motion patterns. Then, we jointly learn the 3D Gaussians, the canonical space locations of control points, and the deformation MLP to reconstruct the appearance, geometry, and dynamics of 3D scenes. During learning, the location and number of control points are adaptively adjusted to accommodate varying motion complexities in different regions, and an ARAP loss following the principle of as rigid as possible is developed to enforce spatial continuity and local rigidity of learned motions. Finally, thanks to the explicit sparse motion representation and its decomposition from appearance, our method can enable user-controlled motion editing while retaining high-fidelity appearances. Extensive experiments demonstrate that our approach outperforms existing approaches on novel view synthesis with a high rendering speed and enables novel appearance-preserved motion editing applications.
Authors: Byung-Hoon Kim, Jungwon Choi, EungGu Yun, Kyungsang Kim, Xiang Li, Juho Lee
Graph Transformers have recently been successful in various graph representation learning tasks, providing a number of advantages over message-passing Graph Neural Networks. Utilizing Graph Transformers for learning the representation of the brain functional connectivity network is also gaining interest. However, studies to date have underlooked the temporal dynamics of functional connectivity, which fluctuates over time. Here, we propose a method for learning the representation of dynamic functional connectivity with Graph Transformers. Specifically, we define the connectome embedding, which holds the position, structure, and time information of the functional connectivity graph, and use Transformers to learn its representation across time. We perform experiments with over 50,000 resting-state fMRI samples obtained from three datasets, which is the largest number of fMRI data used in studies by far. The experimental results show that our proposed method outperforms other competitive baselines in gender classification and age regression tasks based on the functional connectivity extracted from the fMRI data.
Authors: Yunzhe Xue, Olanrewaju Eletta, Justin W. Ady, Nell M. Patel, Advaith Bongu, Usman Roshan
As part of their training all medical students and residents have to pass basic surgical tasks such as knot tying, needle-passing, and suturing. Their assessment is typically performed in the operating room by surgical faculty where mistakes and failure by the student increases the operation time and cost. This evaluation is quantitative and has a low margin of error. Simulation has emerged as a cost effective option but it lacks assessment or requires additional expensive hardware for evaluation. Apps that provide training videos on surgical knot trying are available to students but none have evaluation. We propose a cascaded neural network architecture that evaluates a student's performance just from a video of themselves simulating a surgical knot tying task. Our model converts video frame images into feature vectors with a pre-trained deep convolutional network and then models the sequence of frames with a temporal network. We obtained videos of medical students and residents from the Robert Wood Johnson Hospital performing knot tying on a standardized simulation kit. We manually annotated each video and proceeded to do a five-fold cross-validation study on them. Our model achieves a median precision, recall, and F1-score of 0.71, 0.66, and 0.65 respectively in determining the level of knot related tasks of tying and pushing the knot. Our mean precision score averaged across different probability thresholds is 0.8. Both our F1-score and mean precision score are 8% and 30% higher than that of a recently published study for the same problem. We expect the accuracy of our model to further increase as we add more training videos to the model thus making it a practical solution that students can use to evaluate themselves.
Authors: Vidya Prasad, Chen Zhu-Tian, Anna Vilanova, Hanspeter Pfister, Nicola Pezzotti, Hendrik Strobelt
Diffusion models have garnered significant attention since they can effectively learn complex multivariate Gaussian distributions, resulting in diverse, high-quality outcomes. They introduce Gaussian noise into training data and reconstruct the original data iteratively. Central to this iterative process is a single Unet, adapting across time steps to facilitate generation. Recent work revealed the presence of composition and denoising phases in this generation process, raising questions about the Unets' varying roles. Our study dives into the dynamic behavior of Unets within denoising diffusion probabilistic models (DDPM), focusing on (de)convolutional blocks and skip connections across time steps. We propose an analytical method to systematically assess the impact of time steps and core Unet components on the final output. This method eliminates components to study causal relations and investigate their influence on output changes. The main purpose is to understand the temporal dynamics and identify potential shortcuts during inference. Our findings provide valuable insights into the various generation phases during inference and shed light on the Unets' usage patterns across these phases. Leveraging these insights, we identify redundancies in GLIDE (an improved DDPM) and improve inference time by ~27% with minimal degradation in output quality. Our ultimate goal is to guide more informed optimization strategies for inference and influence new model designs.
Authors: Xinyu Wang, Vinod K. Mishra, C.-C. Jay Kuo
AI algorithms at the edge demand smaller model sizes and lower computational complexity. To achieve these objectives, we adopt a green learning (GL) paradigm rather than the deep learning paradigm. GL has three modules: 1) unsupervised representation learning, 2) supervised feature learning, and 3) supervised decision learning. We focus on the second module in this work. In particular, we derive new discriminant features from proper linear combinations of input features, denoted by x, obtained in the first module. They are called complementary and raw features, respectively. Along this line, we present a novel supervised learning method to generate highly discriminant complementary features based on the least-squares normal transform (LNT). LNT consists of two steps. First, we convert a C-class classification problem to a binary classification problem. The two classes are assigned with 0 and 1, respectively. Next, we formulate a least-squares regression problem from the N-dimensional (N-D) feature space to the 1-D output space, and solve the least-squares normal equation to obtain one N-D normal vector, denoted by a1. Since one normal vector is yielded by one binary split, we can obtain M normal vectors with M splits. Then, Ax is called an LNT of x, where transform matrix A in R^{M by N} by stacking aj^T, j=1, ..., M, and the LNT, Ax, can generate M new features. The newly generated complementary features are shown to be more discriminant than the raw features. Experiments show that the classification performance can be improved by these new features.
Authors: Basudha Pal, Arunkumar Kannan, Ram Prabhakar Kathirvel, Alice J. O'Toole, Rama Chellappa
Diffusion models have achieved great progress in face generation. However, these models amplify the bias in the generation process, leading to an imbalance in distribution of sensitive attributes such as age, gender and race. This paper proposes a novel solution to this problem by balancing the facial attributes of the generated images. We mitigate the bias by localizing the means of the facial attributes in the latent space of the diffusion model using Gaussian mixture models (GMM). Our motivation for choosing GMMs over other clustering frameworks comes from the flexible latent structure of diffusion model. Since each sampling step in diffusion models follows a Gaussian distribution, we show that fitting a GMM model helps us to localize the subspace responsible for generating a specific attribute. Furthermore, our method does not require retraining, we instead localize the subspace on-the-fly and mitigate the bias for generating a fair dataset. We evaluate our approach on multiple face attribute datasets to demonstrate the effectiveness of our approach. Our results demonstrate that our approach leads to a more fair data generation in terms of representational fairness while preserving the quality of generated samples.
Authors: Nannan Li, Qing Liu, Krishna Kumar Singh, Yilin Wang, Jianming Zhang, Bryan A. Plummer, Zhe Lin
Human image editing includes tasks like changing a person's pose, their clothing, or editing the image according to a text prompt. However, prior work often tackles these tasks separately, overlooking the benefit of mutual reinforcement from learning them jointly. In this paper, we propose UniHuman, a unified model that addresses multiple facets of human image editing in real-world settings. To enhance the model's generation quality and generalization capacity, we leverage guidance from human visual encoders and introduce a lightweight pose-warping module that can exploit different pose representations, accommodating unseen textures and patterns. Furthermore, to bridge the disparity between existing human editing benchmarks with real-world data, we curated 400K high-quality human image-text pairs for training and collected 2K human images for out-of-domain testing, both encompassing diverse clothing styles, backgrounds, and age groups. Experiments on both in-domain and out-of-domain test sets demonstrate that UniHuman outperforms task-specific models by a significant margin. In user studies, UniHuman is preferred by the users in an average of 77% of cases.
Authors: Zhangyin Feng, Runyi Hu, Liangxin Liu, Fan Zhang, Duyu Tang, Yong Dai, Xiaocheng Feng, Jiwei Li, Bing Qin, Shuming Shi
Autoregressive and diffusion models drive the recent breakthroughs on text-to-image generation. Despite their huge success of generating high-realistic images, a common shortcoming of these models is their high inference latency - autoregressive models run more than a thousand times successively to produce image tokens and diffusion models convert Gaussian noise into images with many hundreds of denoising steps. In this work, we explore non-autoregressive text-to-image models that efficiently generate hundreds of image tokens in parallel. We develop many model variations with different learning and inference strategies, initialized text encoders, etc. Compared with autoregressive baselines that needs to run one thousand times, our model only runs 16 times to generate images of competitive quality with an order of magnitude lower inference latency. Our non-autoregressive model with 346M parameters generates an image of 256$\times$256 with about one second on one V100 GPU.
Authors: Yujie Li, Xin Yang, Hao Wang, Xiangkun Wang, Tianrui Li
This paper studies the problem of continual learning in an open-world scenario, referred to as Open-world Continual Learning (OwCL). OwCL is increasingly rising while it is highly challenging in two-fold: i) learning a sequence of tasks without forgetting knowns in the past, and ii) identifying unknowns (novel objects/classes) in the future. Existing OwCL methods suffer from the adaptability of task-aware boundaries between knowns and unknowns, and do not consider the mechanism of knowledge transfer. In this work, we propose Pro-KT, a novel prompt-enhanced knowledge transfer model for OwCL. Pro-KT includes two key components: (1) a prompt bank to encode and transfer both task-generic and task-specific knowledge, and (2) a task-aware open-set boundary to identify unknowns in the new tasks. Experimental results using two real-world datasets demonstrate that the proposed Pro-KT outperforms the state-of-the-art counterparts in both the detection of unknowns and the classification of knowns markedly.
Authors: Yuehao Yin, Huiyan Qi, Bin Zhu, Jingjing Chen, Yu-Gang Jiang, Chong-Wah Ngo
Large Multi-modal Models (LMMs) have made impressive progress in many vision-language tasks. Nevertheless, the performance of general LMMs in specific domains is still far from satisfactory. This paper proposes FoodLMM, a versatile food assistant based on LMMs with various capabilities, including food recognition, ingredient recognition, recipe generation, nutrition estimation, food segmentation and multi-round conversation. To facilitate FoodLMM to deal with tasks beyond pure text output, we introduce a series of novel task-specific tokens and heads, enabling the model to predict food nutritional values and multiple segmentation masks. We adopt a two-stage training strategy. In the first stage, we utilize multiple public food benchmarks for multi-task learning by leveraging instruct-following paradigm. In the second stage, we construct a multi-round conversation and a reasoning segmentation datasets to fine-tune the model, enabling it to conduct professional dialogues and generate segmentation masks based on complex reasoning in food domain. Our fine-tuned FoodLMM achieves state-of-the-art results across several food benchmarks. We will make our code, models and datasets publicly available.
Authors: Johann Ostmeyer, Ludovica Schaerf, Pavel Buividovich, Tessa Charles, Eric Postma, Carina Popovici
Previous research has shown that Artificial Intelligence is capable of distinguishing between authentic paintings by a given artist and human-made forgeries with remarkable accuracy, provided sufficient training. However, with the limited amount of existing known forgeries, augmentation methods for forgery detection are highly desirable. In this work, we examine the potential of incorporating synthetic artworks into training datasets to enhance the performance of forgery detection. Our investigation focuses on paintings by Vincent van Gogh, for which we release the first dataset specialized for forgery detection. To reinforce our results, we conduct the same analyses on the artists Amedeo Modigliani and Raphael. We train a classifier to distinguish original artworks from forgeries. For this, we use human-made forgeries and imitations in the style of well-known artists and augment our training sets with images in a similar style generated by Stable Diffusion and StyleGAN. We find that the additional synthetic forgeries consistently improve the detection of human-made forgeries. In addition, we find that, in line with previous research, the inclusion of synthetic forgeries in the training also enables the detection of AI-generated forgeries, especially if created using a similar generator.
Authors: Tin Nguyen, Anh Nguyen
Traditional bird classifiers mostly rely on the visual characteristics of birds. Some prior works even train classifiers to be invariant to the background, completely discarding the living environment of birds. Instead, we are the first to explore integrating habitat information, one of the four major cues for identifying birds by ornithologists, into modern bird classifiers. We focus on two leading model types: (1) CNNs and ViTs trained on the downstream bird datasets; and (2) original, multi-modal CLIP. Training CNNs and ViTs with habitat-augmented data results in an improvement of up to +0.83 and +0.23 points on NABirds and CUB-200, respectively. Similarly, adding habitat descriptors to the prompts for CLIP yields a substantial accuracy boost of up to +0.99 and +1.1 points on NABirds and CUB-200, respectively. We find consistent accuracy improvement after integrating habitat features into the image augmentation process and into the textual descriptors of vision-language CLIP classifiers. Code is available at: https://anonymous.4open.science/r/reasoning-8B7E/.
Authors: Mingyuan Zhang, Huirong Li, Zhongang Cai, Jiawei Ren, Lei Yang, Ziwei Liu
Text-driven motion generation has achieved substantial progress with the emergence of diffusion models. However, existing methods still struggle to generate complex motion sequences that correspond to fine-grained descriptions, depicting detailed and accurate spatio-temporal actions. This lack of fine controllability limits the usage of motion generation to a larger audience. To tackle these challenges, we present FineMoGen, a diffusion-based motion generation and editing framework that can synthesize fine-grained motions, with spatial-temporal composition to the user instructions. Specifically, FineMoGen builds upon diffusion model with a novel transformer architecture dubbed Spatio-Temporal Mixture Attention (SAMI). SAMI optimizes the generation of the global attention template from two perspectives: 1) explicitly modeling the constraints of spatio-temporal composition; and 2) utilizing sparsely-activated mixture-of-experts to adaptively extract fine-grained features. To facilitate a large-scale study on this new fine-grained motion generation task, we contribute the HuMMan-MoGen dataset, which consists of 2,968 videos and 102,336 fine-grained spatio-temporal descriptions. Extensive experiments validate that FineMoGen exhibits superior motion generation quality over state-of-the-art methods. Notably, FineMoGen further enables zero-shot motion editing capabilities with the aid of modern large language models (LLM), which faithfully manipulates motion sequences with fine-grained instructions. Project Page: https://mingyuan-zhang.github.io/projects/FineMoGen.html
Authors: Saarthak Kapse, Pushpak Pati, Srijan Das, Jingwei Zhang, Chao Chen, Maria Vakalopoulou, Joel Saltz, Dimitris Samaras, Rajarsi R. Gupta, Prateek Prasanna
Introducing interpretability and reasoning into Multiple Instance Learning (MIL) methods for Whole Slide Image (WSI) analysis is challenging, given the complexity of gigapixel slides. Traditionally, MIL interpretability is limited to identifying salient regions deemed pertinent for downstream tasks, offering little insight to the end-user (pathologist) regarding the rationale behind these selections. To address this, we propose Self-Interpretable MIL (SI-MIL), a method intrinsically designed for interpretability from the very outset. SI-MIL employs a deep MIL framework to guide an interpretable branch grounded on handcrafted pathological features, facilitating linear predictions. Beyond identifying salient regions, SI-MIL uniquely provides feature-level interpretations rooted in pathological insights for WSIs. Notably, SI-MIL, with its linear prediction constraints, challenges the prevalent myth of an inevitable trade-off between model interpretability and performance, demonstrating competitive results compared to state-of-the-art methods on WSI-level prediction tasks across three cancer types. In addition, we thoroughly benchmark the local- and global-interpretability of SI-MIL in terms of statistical analysis, a domain expert study, and desiderata of interpretability, namely, user-friendliness and faithfulness.
Authors: Zhangyang Qi, Ye Fang, Mengchen Zhang, Zeyi Sun, Tong Wu, Ziwei Liu, Dahua Lin, Jiaqi Wang, Hengshuang Zhao
The rapidly evolving sector of Multi-modal Large Language Models (MLLMs) is at the forefront of integrating linguistic and visual processing in artificial intelligence. This paper presents an in-depth comparative study of two pioneering models: Google's Gemini and OpenAI's GPT-4V(ision). Our study involves a multi-faceted evaluation of both models across key dimensions such as Vision-Language Capability, Interaction with Humans, Temporal Understanding, and assessments in both Intelligence and Emotional Quotients. The core of our analysis delves into the distinct visual comprehension abilities of each model. We conducted a series of structured experiments to evaluate their performance in various industrial application scenarios, offering a comprehensive perspective on their practical utility. We not only involve direct performance comparisons but also include adjustments in prompts and scenarios to ensure a balanced and fair analysis. Our findings illuminate the unique strengths and niches of both models. GPT-4V distinguishes itself with its precision and succinctness in responses, while Gemini excels in providing detailed, expansive answers accompanied by relevant imagery and links. These understandings not only shed light on the comparative merits of Gemini and GPT-4V but also underscore the evolving landscape of multimodal foundation models, paving the way for future advancements in this area. After the comparison, we attempted to achieve better results by combining the two models. Finally, We would like to express our profound gratitude to the teams behind GPT-4V and Gemini for their pioneering contributions to the field. Our acknowledgments are also extended to the comprehensive qualitative analysis presented in 'Dawn' by Yang et al. This work, with its extensive collection of image samples, prompts, and GPT-4V-related results, provided a foundational basis for our analysis.
Authors: Snehal Singh Tomar, A.N. Rajagopalan
With the metaverse slowly becoming a reality and given the rapid pace of developments toward the creation of digital humans, the need for a principled style editing pipeline for human faces is bound to increase manifold. We cater to this need by introducing the Latents2Semantics Autoencoder (L2SAE), a Generative Autoencoder model that facilitates highly localized editing of style attributes of several Regions of Interest (ROIs) in face images. The L2SAE learns separate latent representations for encoded images' structure and style information. Thus, allowing for structure-preserving style editing of the chosen ROIs. The encoded structure representation is a multichannel 2D tensor with reduced spatial dimensions, which captures both local and global structure properties. The style representation is a 1D tensor that captures global style attributes. In our framework, we slice the structure representation to build strong and disentangled correspondences with different ROIs. Consequentially, style editing of the chosen ROIs amounts to a simple combination of (a) the ROI-mask generated from the sliced structure representation and (b) the decoded image with global style changes, generated from the manipulated (using Gaussian noise) global style and unchanged structure tensor. Style editing sans additional human supervision is a significant win over SOTA style editing pipelines because most existing works require additional human effort (supervision) post-training for attributing semantic meaning to style edits. We also do away with iterative-optimization-based inversion or determining controllable latent directions post-training, which requires additional computationally expensive operations. We provide qualitative and quantitative results for the same over multiple applications, such as selective style editing and swapping using test images sampled from several datasets.
Authors: Haozhan Shen, Tiancheng Zhao, Mingwei Zhu, Jianwei Yin
Visual grounding, a crucial vision-language task involving the understanding of the visual context based on the query expression, necessitates the model to capture the interactions between objects, as well as various spatial and attribute information. However, the annotation data of visual grounding task is limited due to its time-consuming and labor-intensive annotation process, resulting in the trained models being constrained from generalizing its capability to a broader domain. To address this challenge, we propose GroundVLP, a simple yet effective zero-shot method that harnesses visual grounding ability from the existing models trained from image-text pairs and pure object detection data, both of which are more conveniently obtainable and offer a broader domain compared to visual grounding annotation data. GroundVLP proposes a fusion mechanism that combines the heatmap from GradCAM and the object proposals of open-vocabulary detectors. We demonstrate that the proposed method significantly outperforms other zero-shot methods on RefCOCO/+/g datasets, surpassing prior zero-shot state-of-the-art by approximately 28\% on the test split of RefCOCO and RefCOCO+. Furthermore, GroundVLP performs comparably to or even better than some non-VLP-based supervised models on the Flickr30k entities dataset. Our code is available at https://github.com/om-ai-lab/GroundVLP.
Authors: HyunJun Jung, Nikolas Brasch, Jifei Song, Eduardo Perez-Pellitero, Yiren Zhou, Zhihao Li, Nassir Navab, Benjamin Busam
Recent advances in neural radiance fields enable novel view synthesis of photo-realistic images in dynamic settings, which can be applied to scenarios with human animation. Commonly used implicit backbones to establish accurate models, however, require many input views and additional annotations such as human masks, UV maps and depth maps. In this work, we propose ParDy-Human (Parameterized Dynamic Human Avatar), a fully explicit approach to construct a digital avatar from as little as a single monocular sequence. ParDy-Human introduces parameter-driven dynamics into 3D Gaussian Splatting where 3D Gaussians are deformed by a human pose model to animate the avatar. Our method is composed of two parts: A first module that deforms canonical 3D Gaussians according to SMPL vertices and a consecutive module that further takes their designed joint encodings and predicts per Gaussian deformations to deal with dynamics beyond SMPL vertex deformations. Images are then synthesized by a rasterizer. ParDy-Human constitutes an explicit model for realistic dynamic human avatars which requires significantly fewer training views and images. Our avatars learning is free of additional annotations such as masks and can be trained with variable backgrounds while inferring full-resolution images efficiently even on consumer hardware. We provide experimental evidence to show that ParDy-Human outperforms state-of-the-art methods on ZJU-MoCap and THUman4.0 datasets both quantitatively and visually.
Authors: Zhiyuan Li, Hailong Li, Anca L. Ralescu, Jonathan R. Dillman, Mekibib Altaye, Kim M. Cecil, Nehal A. Parikh, Lili He
The integration of different imaging modalities, such as structural, diffusion tensor, and functional magnetic resonance imaging, with deep learning models has yielded promising outcomes in discerning phenotypic characteristics and enhancing disease diagnosis. The development of such a technique hinges on the efficient fusion of heterogeneous multimodal features, which initially reside within distinct representation spaces. Naively fusing the multimodal features does not adequately capture the complementary information and could even produce redundancy. In this work, we present a novel joint self-supervised and supervised contrastive learning method to learn the robust latent feature representation from multimodal MRI data, allowing the projection of heterogeneous features into a shared common space, and thereby amalgamating both complementary and analogous information across various modalities and among similar subjects. We performed a comparative analysis between our proposed method and alternative deep multimodal learning approaches. Through extensive experiments on two independent datasets, the results demonstrated that our method is significantly superior to several other deep multimodal learning methods in predicting abnormal neurodevelopment. Our method has the capability to facilitate computer-aided diagnosis within clinical practice, harnessing the power of multimodal data.
Authors: Binbin Xiang, Maciej Wielgosz, Theodora Kontogianni, Torben Peters, Stefano Puliti, Rasmus Astrup, Konrad Schindler
Detailed forest inventories are critical for sustainable and flexible management of forest resources, to conserve various ecosystem services. Modern airborne laser scanners deliver high-density point clouds with great potential for fine-scale forest inventory and analysis, but automatically partitioning those point clouds into meaningful entities like individual trees or tree components remains a challenge. The present study aims to fill this gap and introduces a deep learning framework that is able to perform such a segmentation across diverse forest types and geographic regions. From the segmented data, we then derive relevant biophysical parameters of individual trees as well as stands. The system has been tested on FOR-Instance, a dataset of point clouds that have been acquired in five different countries using surveying drones. The segmentation back-end achieves over 85% F-score for individual trees, respectively over 73% mean IoU across five semantic categories: ground, low vegetation, stems, live branches and dead branches. Building on the segmentation results our pipeline then densely calculates biophysical features of each individual tree (height, crown diameter, crown volume, DBH, and location) and properties per stand (digital terrain model and stand density). Especially crown-related features are in most cases retrieved with high accuracy, whereas the estimates for DBH and location are less reliable, due to the airborne scanning setup.
Authors: Nikhil Mehta, Kevin J Liang, Jing Huang, Fu-Jen Chu, Li Yin, Tal Hassner
Out-of-distribution (OOD) detection is an important topic for real-world machine learning systems, but settings with limited in-distribution samples have been underexplored. Such few-shot OOD settings are challenging, as models have scarce opportunities to learn the data distribution before being tasked with identifying OOD samples. Indeed, we demonstrate that recent state-of-the-art OOD methods fail to outperform simple baselines in the few-shot setting. We thus propose a hypernetwork framework called HyperMix, using Mixup on the generated classifier parameters, as well as a natural out-of-episode outlier exposure technique that does not require an additional outlier dataset. We conduct experiments on CIFAR-FS and MiniImageNet, significantly outperforming other OOD methods in the few-shot regime.
Authors: Nikolaos Louloudakis, Perry Gibson, José Cano, Ajitha Rajan
Converting deep learning models between frameworks is a common step to maximize model compatibility across devices and leverage optimization features that may be exclusively provided in one deep learning framework. However, this conversion process may be riddled with bugs, making the converted models either undeployable or problematic, considerably degrading their prediction correctness.
We propose an automated approach for fault localization and repair, Fix-Con, during model conversion between deep learning frameworks. Fix-Con is capable of detecting and fixing faults introduced in model input, parameters, hyperparameters, and the model graph during conversion.
Fix-Con uses a set of fault types mined from surveying conversion issues raised to localize potential conversion faults in the converted target model, and then repairs them appropriately, e.g. replacing the parameters of the target model with those from the source model. This is done iteratively for every image in the dataset with output label differences between the source model and the converted target model until all differences are resolved. We evaluate the effectiveness of Fix-Con in fixing model conversion bugs of three widely used image recognition models converted across four different deep learning frameworks. Overall, Fix-Con was able to either completely repair, or significantly improve the performance of 14 out of the 15 erroneous conversion cases.
Authors: Wassim Kabbani, Christoph Busch, Kiran Raja
Face image quality assessment (FIQA) is crucial for obtaining good face recognition performance. FIQA algorithms should be robust and insensitive to demographic factors. The eye sclera has a consistent whitish color in all humans regardless of their age, ethnicity and skin-tone. This work proposes a robust sclera segmentation method that is suitable for face images in the enrolment and the border control face recognition scenarios. It shows how the statistical analysis of the sclera pixels produces features that are invariant to skin-tone, age and ethnicity and thus can be incorporated into FIQA algorithms to make them agnostic to demographic factors.
Authors: Benjamin Scellier, Maxence Ernoult, Jack Kendall, Suhas Kumar
Energy-based learning algorithms have recently gained a surge of interest due to their compatibility with analog (post-digital) hardware. Existing algorithms include contrastive learning (CL), equilibrium propagation (EP) and coupled learning (CpL), all consisting in contrasting two states, and differing in the type of perturbation used to obtain the second state from the first one. However, these algorithms have never been explicitly compared on equal footing with same models and datasets, making it difficult to assess their scalability and decide which one to select in practice. In this work, we carry out a comparison of seven learning algorithms, namely CL and different variants of EP and CpL depending on the signs of the perturbations. Specifically, using these learning algorithms, we train deep convolutional Hopfield networks (DCHNs) on five vision tasks (MNIST, F-MNIST, SVHN, CIFAR-10 and CIFAR-100). We find that, while all algorithms yield comparable performance on MNIST, important differences in performance arise as the difficulty of the task increases. Our key findings reveal that negative perturbations are better than positive ones, and highlight the centered variant of EP (which uses two perturbations of opposite sign) as the best-performing algorithm. We also endorse these findings with theoretical arguments. Additionally, we establish new SOTA results with DCHNs on all five datasets, both in performance and speed. In particular, our DCHN simulations are 13.5 times faster with respect to Laborieux et al. (2021), which we achieve thanks to the use of a novel energy minimisation algorithm based on asynchronous updates, combined with reduced precision (16 bits).
Authors: Wassim Kabbani, Marcel Grimmer, Christoph Busch
Generative Adversarial Networks (GANs) have witnessed significant advances in recent years, generating increasingly higher quality images, which are non-distinguishable from real ones. Recent GANs have proven to encode features in a disentangled latent space, enabling precise control over various semantic attributes of the generated facial images such as pose, illumination, or gender. GAN inversion, which is projecting images into the latent space of a GAN, opens the door for the manipulation of facial semantics of real face images. This is useful for numerous applications such as evaluating the performance of face recognition systems. In this work, EGAIN, an architecture for constructing GAN inversion models, is presented. This architecture explicitly addresses some of the shortcomings in previous GAN inversion models. A specific model with the same name, egain, based on this architecture is also proposed, demonstrating superior reconstruction quality over state-of-the-art models, and illustrating the validity of the EGAIN architecture.
Authors: Yang You, Kai Xiong, Zhening Yang, Zhengxiang Huang, Junwei Zhou, Ruoxi Shi, Zhou Fang, Adam W. Harley, Cewu Lu
Pose estimation is a crucial task in computer vision, enabling tracking and manipulating objects in images or videos. While several datasets exist for pose estimation, there is a lack of large-scale datasets specifically focusing on cluttered scenes with occlusions. This limitation is a bottleneck in the development and evaluation of pose estimation methods, particularly toward the goal of real-world application in environments where occlusions are common. Addressing this, we introduce PACE (Pose Annotations in Cluttered Environments), a large-scale benchmark designed to advance the development and evaluation of pose estimation methods in cluttered scenarios. PACE encompasses 54,945 frames with 257,673 annotations across 300 videos, covering 576 objects from 44 categories and featuring a mix of rigid and articulated items in cluttered scenes. To annotate the real-world data efficiently, we developed an innovative annotation system utilizing a calibrated 3-camera setup. We test state-of-the-art algorithms in PACE along two tracks: pose estimation, and object pose tracking, revealing the benchmark's challenges and research opportunities. We plan to release PACE as a public evaluation benchmark, along the annotations tools we developed, to stimulate further advancements in the field. Our code and data is available on https://github.com/qq456cvb/PACE.
Authors: Shujuan Li, Junsheng Zhou, Baorui Ma, Yu-Shen Liu, Zhizhong Han
Point cloud upsampling aims to generate dense and uniformly distributed point sets from a sparse point cloud, which plays a critical role in 3D computer vision. Previous methods typically split a sparse point cloud into several local patches, upsample patch points, and merge all upsampled patches. However, these methods often produce holes, outliers or nonuniformity due to the splitting and merging process which does not maintain consistency among local patches. To address these issues, we propose a novel approach that learns an unsigned distance field guided by local priors for point cloud upsampling. Specifically, we train a local distance indicator (LDI) that predicts the unsigned distance from a query point to a local implicit surface. Utilizing the learned LDI, we learn an unsigned distance field to represent the sparse point cloud with patch consistency. At inference time, we randomly sample queries around the sparse point cloud, and project these query points onto the zero-level set of the learned implicit field to generate a dense point cloud. We justify that the implicit field is naturally continuous, which inherently enables the application of arbitrary-scale upsampling without necessarily retraining for various scales. We conduct comprehensive experiments on both synthetic data and real scans, and report state-of-the-art results under widely used benchmarks.
Authors: Gabe Guo, Judah Goldfeder, Ling Lan, Aniv Ray, Albert Hanming Yang, Boyuan Chen, Simon JL Billinge, Hod Lipson
The revolution in materials in the past century was built on a knowledge of the atomic arrangements and the structure-property relationship. The sine qua non for obtaining quantitative structural information is single crystal crystallography. However, increasingly we need to solve structures in cases where the information content in our input signal is significantly degraded, for example, due to orientational averaging of grains, finite size effects due to nanostructure, and mixed signals due to sample heterogeneity. Understanding the structure property relationships in such situations is, if anything, more important and insightful, yet we do not have robust approaches for accomplishing it. In principle, machine learning (ML) and deep learning (DL) are promising approaches since they augment information in the degraded input signal with prior knowledge learned from large databases of already known structures. Here we present a novel ML approach, a variational query-based multi-branch deep neural network that has the promise to be a robust but general tool to address this problem end-to-end. We demonstrate the approach on computed powder x-ray diffraction (PXRD), along with partial chemical composition information, as input. We choose as a structural representation a modified electron density we call the Cartesian mapped electron density (CMED), that straightforwardly allows our ML model to learn material structures across different chemistries, symmetries and crystal systems. When evaluated on theoretically simulated data for the cubic and trigonal crystal systems, the system achieves up to $93.4\%$ average similarity with the ground truth on unseen materials, both with known and partially-known chemical composition information, showing great promise for successful structure solution even from degraded and incomplete input data.
Authors: Changsong Lei, Mengfei Xia, Shaofeng Wang, Yaqian Liang, Ran Yi, Yuhui Wen, Yongjin Liu
Tooth arrangement is a crucial step in orthodontics treatment, in which aligning teeth could improve overall well-being, enhance facial aesthetics, and boost self-confidence. To improve the efficiency of tooth arrangement and minimize errors associated with unreasonable designs by inexperienced practitioners, some deep learning-based tooth arrangement methods have been proposed. Currently, most existing approaches employ MLPs to model the nonlinear relationship between tooth features and transformation matrices to achieve tooth arrangement automatically. However, the limited datasets (which to our knowledge, have not been made public) collected from clinical practice constrain the applicability of existing methods, making them inadequate for addressing diverse malocclusion issues. To address this challenge, we propose a general tooth arrangement neural network based on the diffusion probabilistic model. Conditioned on the features extracted from the dental model, the diffusion probabilistic model can learn the distribution of teeth transformation matrices from malocclusion to normal occlusion by gradually denoising from a random variable, thus more adeptly managing real orthodontic data. To take full advantage of effective features, we exploit both mesh and point cloud representations by designing different encoding networks to extract the tooth (local) and jaw (global) features, respectively. In addition to traditional metrics ADD, PA-ADD, CSA, and ME_{rot}, we propose a new evaluation metric based on dental arch curves to judge whether the generated teeth meet the individual normal occlusion. Experimental results demonstrate that our proposed method achieves state-of-the-art tooth alignment results and satisfactory occlusal relationships between dental arches. We will publish the code and dataset.
Authors: Ruiping Liu, Kailun Yang, Alina Roitberg, Jiaming Zhang, Kunyu Peng, Huayao Liu, Yaonan Wang, Rainer Stiefelhagen
Semantic segmentation benchmarks in the realm of autonomous driving are dominated by large pre-trained transformers, yet their widespread adoption is impeded by substantial computational costs and prolonged training durations. To lift this constraint, we look at efficient semantic segmentation from a perspective of comprehensive knowledge distillation and consider to bridge the gap between multi-source knowledge extractions and transformer-specific patch embeddings. We put forward the Transformer-based Knowledge Distillation (TransKD) framework which learns compact student transformers by distilling both feature maps and patch embeddings of large teacher transformers, bypassing the long pre-training process and reducing the FLOPs by >85.0%. Specifically, we propose two fundamental and two optimization modules: (1) Cross Selective Fusion (CSF) enables knowledge transfer between cross-stage features via channel attention and feature map distillation within hierarchical transformers; (2) Patch Embedding Alignment (PEA) performs dimensional transformation within the patchifying process to facilitate the patch embedding distillation; (3) Global-Local Context Mixer (GL-Mixer) extracts both global and local information of a representative embedding; (4) Embedding Assistant (EA) acts as an embedding method to seamlessly bridge teacher and student models with the teacher's number of channels. Experiments on Cityscapes, ACDC, NYUv2, and Pascal VOC2012 datasets show that TransKD outperforms state-of-the-art distillation frameworks and rivals the time-consuming pre-training method. The source code is publicly available at https://github.com/RuipingL/TransKD.
Authors: Haotian Ju, Dongyue Li, Hongyang R. Zhang
We consider fine-tuning a pretrained deep neural network on a target task. We study the generalization properties of fine-tuning to understand the problem of overfitting, which has often been observed (e.g., when the target dataset is small or when the training labels are noisy). Existing generalization measures for deep networks depend on notions such as distance from the initialization (i.e., the pretrained network) of the fine-tuned model and noise stability properties of deep networks. This paper identifies a Hessian-based distance measure through PAC-Bayesian analysis, which is shown to correlate well with observed generalization gaps of fine-tuned models. Theoretically, we prove Hessian distance-based generalization bounds for fine-tuned models. We also describe an extended study of fine-tuning against label noise, where overfitting remains a critical problem. We present an algorithm and a generalization error guarantee for this algorithm under a class conditional independent noise model. Empirically, we observe that the Hessian-based distance measure can match the scale of the observed generalization gap of fine-tuned models in practice. We also test our algorithm on several image classification tasks with noisy training labels, showing gains over prior methods and decreases in the Hessian distance measure of the fine-tuned model.
Authors: Yan Lin, Huaiyu Wan, Shengnan Guo, Jilin Hu, Christian S. Jensen, Youfang Lin
Spatio-temporal trajectories provide valuable information about movement and travel behavior, enabling various downstream tasks that in turn power real-world applications. Learning trajectory embeddings can improve task performance but may incur high computational costs and face limited training data availability. Pre-training learns generic embeddings by means of specially constructed pretext tasks that enable learning from unlabeled data. Existing pre-training methods face (i) difficulties in learning general embeddings due to biases towards certain downstream tasks incurred by the pretext tasks, (ii) limitations in capturing both travel semantics and spatio-temporal correlations, and (iii) the complexity of long, irregularly sampled trajectories.
To tackle these challenges, we propose Maximum Multi-view Trajectory Entropy Coding (MMTEC) for learning general and comprehensive trajectory embeddings. We introduce a pretext task that reduces biases in pre-trained trajectory embeddings, yielding embeddings that are useful for a wide variety of downstream tasks. We also propose an attention-based discrete encoder and a NeuralCDE-based continuous encoder that extract and represent travel behavior and continuous spatio-temporal correlations from trajectories in embeddings, respectively. Extensive experiments on two real-world datasets and three downstream tasks offer insight into the design properties of our proposal and indicate that it is capable of outperforming existing trajectory embedding methods.
Authors: Jacob Krantz, Shurjo Banerjee, Wang Zhu, Jason Corso, Peter Anderson, Stefan Lee, Jesse Thomason
We present Iterative Vision-and-Language Navigation (IVLN), a paradigm for evaluating language-guided agents navigating in a persistent environment over time. Existing Vision-and-Language Navigation (VLN) benchmarks erase the agent's memory at the beginning of every episode, testing the ability to perform cold-start navigation with no prior information. However, deployed robots occupy the same environment for long periods of time. The IVLN paradigm addresses this disparity by training and evaluating VLN agents that maintain memory across tours of scenes that consist of up to 100 ordered instruction-following Room-to-Room (R2R) episodes, each defined by an individual language instruction and a target path. We present discrete and continuous Iterative Room-to-Room (IR2R) benchmarks comprising about 400 tours each in 80 indoor scenes. We find that extending the implicit memory of high-performing transformer VLN agents is not sufficient for IVLN, but agents that build maps can benefit from environment persistence, motivating a renewed focus on map-building agents in VLN.
Authors: Yun Liu, Xiaomeng Xu, Weihang Chen, Haocheng Yuan, He Wang, Jing Xu, Rui Chen, Li Yi
When manipulating an object to accomplish complex tasks, humans rely on both vision and touch to keep track of the object's 6D pose. However, most existing object pose tracking systems in robotics rely exclusively on visual signals, which hinder a robot's ability to manipulate objects effectively. To address this limitation, we introduce TEG-Track, a tactile-enhanced 6D pose tracking system that can track previously unseen objects held in hand. From consecutive tactile signals, TEG-Track optimizes object velocities from marker flows when slippage does not occur, or regresses velocities using a slippage estimation network when slippage is detected. The estimated object velocities are integrated into a geometric-kinematic optimization scheme to enhance existing visual pose trackers. To evaluate our method and to facilitate future research, we construct a real-world dataset for visual-tactile in-hand object pose tracking. Experimental results demonstrate that TEG-Track consistently enhances state-of-the-art generalizable 6D pose trackers in synthetic and real-world scenarios. Our code and dataset are available at https://github.com/leolyliu/TEG-Track.
Authors: Pritam Sarkar, Ali Etemad
We present XKD, a novel self-supervised framework to learn meaningful representations from unlabelled videos. XKD is trained with two pseudo objectives. First, masked data reconstruction is performed to learn modality-specific representations from audio and visual streams. Next, self-supervised cross-modal knowledge distillation is performed between the two modalities through a teacher-student setup to learn complementary information. We introduce a novel domain alignment strategy to tackle domain discrepancy between audio and visual modalities enabling effective cross-modal knowledge distillation. Additionally, to develop a general-purpose network capable of handling both audio and visual streams, modality-agnostic variants of XKD are introduced, which use the same pretrained backbone for different audio and visual tasks. Our proposed cross-modal knowledge distillation improves video action classification by $8\%$ to $14\%$ on UCF101, HMDB51, and Kinetics400. Additionally, XKD improves multimodal action classification by $5.5\%$ on Kinetics-Sound. XKD shows state-of-the-art performance in sound classification on ESC50, achieving top-1 accuracy of $96.5\%$.
Authors: Simon Klenk, David Bonello, Lukas Koestler, Nikita Araslanov, Daniel Cremers
Event cameras asynchronously capture brightness changes with low latency, high temporal resolution, and high dynamic range. However, annotation of event data is a costly and laborious process, which limits the use of deep learning methods for classification and other semantic tasks with the event modality. To reduce the dependency on labeled event data, we introduce Masked Event Modeling (MEM), a self-supervised framework for events. Our method pretrains a neural network on unlabeled events, which can originate from any event camera recording. Subsequently, the pretrained model is finetuned on a downstream task, leading to a consistent improvement of the task accuracy. For example, our method reaches state-of-the-art classification accuracy across three datasets, N-ImageNet, N-Cars, and N-Caltech101, increasing the top-1 accuracy of previous work by significant margins. When tested on real-world event data, MEM is even superior to supervised RGB-based pretraining. The models pretrained with MEM are also label-efficient and generalize well to the dense task of semantic image segmentation.
Authors: Junde Wu, Wei Ji, Huazhu Fu, Min Xu, Yueming Jin, Yanwu Xu
The Diffusion Probabilistic Model (DPM) has recently gained popularity in the field of computer vision, thanks to its image generation applications, such as Imagen, Latent Diffusion Models, and Stable Diffusion, which have demonstrated impressive capabilities and sparked much discussion within the community. Recent investigations have further unveiled the utility of DPM in the domain of medical image analysis, as underscored by the commendable performance exhibited by the medical image segmentation model across various tasks. Although these models were originally underpinned by a UNet architecture, there exists a potential avenue for enhancing their performance through the integration of vision transformer mechanisms. However, we discovered that simply combining these two models resulted in subpar performance. To effectively integrate these two cutting-edge techniques for the Medical image segmentation, we propose a novel Transformer-based Diffusion framework, called MedSegDiff-V2. We verify its effectiveness on 20 medical image segmentation tasks with different image modalities. Through comprehensive evaluation, our approach demonstrates superiority over prior state-of-the-art (SOTA) methodologies. Code is released at https://github.com/KidsWithTokens/MedSegDiff
Authors: Jian Zhu, Hanli Wang, Miaojing Shi
The visual commonsense reasoning (VCR) task is to choose an answer and provide a justifying rationale based on the given image and textural question. Representative works first recognize objects in images and then associate them with key words in texts. However, existing approaches do not consider exact positions of objects in a human-like three-dimensional (3D) manner, making them incompetent to accurately distinguish objects and understand visual relation. Recently, multi-modal large language models (MLLMs) have been used as powerful tools for several multi-modal tasks but not for VCR yet, which requires elaborate reasoning on specific visual objects referred by texts. In light of the above, an MLLM enhanced pseudo 3D perception framework is designed for VCR. Specifically, we first demonstrate that the relation between objects is relevant to object depths in images, and hence introduce object depth into VCR frameworks to infer 3D positions of objects in images. Then, a depth-aware Transformer is proposed to encode depth differences between objects into the attention mechanism of Transformer to discriminatively associate objects with visual scenes guided by depth. To further associate the answer with the depth of visual scene, each word in the answer is tagged with a pseudo depth to realize depth-aware association between answer words and objects. On the other hand, BLIP-2 as an MLLM is employed to process images and texts, and the referring expressions in texts involving specific visual objects are modified with linguistic object labels to serve as comprehensible MLLM inputs. Finally, a parameter optimization technique is devised to fully consider the quality of data batches based on multi-level reasoning confidence. Experiments on the VCR dataset demonstrate the superiority of the proposed framework over state-of-the-art approaches.
Authors: Sifan Song, Jinfeng Wang, Zilong Wang, Jionglong Su, Xiaowei Ding, Kang Dang
Accurate fovea localization is essential for analyzing retinal diseases to prevent irreversible vision loss. While current deep learning-based methods outperform traditional ones, they still face challenges such as the lack of local anatomical landmarks around the fovea, the inability to robustly handle diseased retinal images, and the variations in image conditions. In this paper, we propose a novel transformer-based architecture called DualStreamFoveaNet (DSFN) for multi-cue fusion. This architecture explicitly incorporates long-range connections and global features using retina and vessel distributions for robust fovea localization. We introduce a spatial attention mechanism in the dual-stream encoder to extract and fuse self-learned anatomical information, focusing more on features distributed along blood vessels and significantly reducing computational costs by decreasing token numbers. Our extensive experiments show that the proposed architecture achieves state-of-the-art performance on two public datasets and one large-scale private dataset. Furthermore, we demonstrate that the DSFN is more robust on both normal and diseased retina images and has better generalization capacity in cross-dataset experiments.
Authors: Kunyu Peng, David Schneider, Alina Roitberg, Kailun Yang, Jiaming Zhang, Chen Deng, Kaiyu Zhang, M. Saquib Sarfraz, Rainer Stiefelhagen
In this paper, we tackle the new task of video-based Activated Muscle Group Estimation (AMGE) aiming at identifying active muscle regions during physical activity in the wild. To this intent, we provide the MuscleMap dataset featuring >15K video clips with 135 different activities and 20 labeled muscle groups. This dataset opens the vistas to multiple video-based applications in sports and rehabilitation medicine under flexible environment constraints. The proposed MuscleMap dataset is constructed with YouTube videos, specifically targeting High-Intensity Interval Training (HIIT) physical exercise in the wild. To make the AMGE model applicable in real-life situations, it is crucial to ensure that the model can generalize well to numerous types of physical activities not present during training and involving new combinations of activated muscles. To achieve this, our benchmark also covers an evaluation setting where the model is exposed to activity types excluded from the training set. Our experiments reveal that the generalizability of existing architectures adapted for the AMGE task remains a challenge. Therefore, we also propose a new approach, TransM3E, which employs a multi-modality feature fusion mechanism between both the video transformer model and the skeleton-based graph convolution model with novel cross-modal knowledge distillation executed on multi-classification tokens. The proposed method surpasses all popular video classification models when dealing with both, previously seen and new types of physical activities. The contributed dataset and code are made publicly available at https://github.com/KPeng9510/MuscleMap.
Authors: Chen Song, Chandrajit Bajaj, Qixing Huang
We present DeblurSR, a novel motion deblurring approach that converts a blurry image into a sharp video. DeblurSR utilizes event data to compensate for motion ambiguities and exploits the spiking representation to parameterize the sharp output video as a mapping from time to intensity. Our key contribution, the Spiking Representation (SR), is inspired by the neuromorphic principles determining how biological neurons communicate with each other in living organisms. We discuss why the spikes can represent sharp edges and how the spiking parameters are interpreted from the neuromorphic perspective. DeblurSR has higher output quality and requires fewer computing resources than state-of-the-art event-based motion deblurring methods. We additionally show that our approach easily extends to video super-resolution when combined with recent advances in implicit neural representation. The implementation and animated visualization of DeblurSR are available at https://github.com/chensong1995/DeblurSR.
Authors: Wulian Yun, Mengshi Qi, Chuanming Wang, Huadong Ma
Weakly-supervised temporal action localization aims to locate action regions and identify action categories in untrimmed videos simultaneously by taking only video-level labels as the supervision. Pseudo label generation is a promising strategy to solve the challenging problem, but the current methods ignore the natural temporal structure of the video that can provide rich information to assist such a generation process. In this paper, we propose a novel weakly-supervised temporal action localization method by inferring salient snippet-feature. First, we design a saliency inference module that exploits the variation relationship between temporal neighbor snippets to discover salient snippet-features, which can reflect the significant dynamic change in the video. Secondly, we introduce a boundary refinement module that enhances salient snippet-features through the information interaction unit. Then, a discrimination enhancement module is introduced to enhance the discriminative nature of snippet-features. Finally, we adopt the refined snippet-features to produce high-fidelity pseudo labels, which could be used to supervise the training of the action localization network. Extensive experiments on two publicly available datasets, i.e., THUMOS14 and ActivityNet v1.3, demonstrate our proposed method achieves significant improvements compared to the state-of-the-art methods.
Authors: Youngseok Kim, Juyeb Shin, Sanmin Kim, In-Jae Lee, Jun Won Choi, Dongsuk Kum
Autonomous driving requires an accurate and fast 3D perception system that includes 3D object detection, tracking, and segmentation. Although recent low-cost camera-based approaches have shown promising results, they are susceptible to poor illumination or bad weather conditions and have a large localization error. Hence, fusing camera with low-cost radar, which provides precise long-range measurement and operates reliably in all environments, is promising but has not yet been thoroughly investigated. In this paper, we propose Camera Radar Net (CRN), a novel camera-radar fusion framework that generates a semantically rich and spatially accurate bird's-eye-view (BEV) feature map for various tasks. To overcome the lack of spatial information in an image, we transform perspective view image features to BEV with the help of sparse but accurate radar points. We further aggregate image and radar feature maps in BEV using multi-modal deformable attention designed to tackle the spatial misalignment between inputs. CRN with real-time setting operates at 20 FPS while achieving comparable performance to LiDAR detectors on nuScenes, and even outperforms at a far distance on 100m setting. Moreover, CRN with offline setting yields 62.4% NDS, 57.5% mAP on nuScenes test set and ranks first among all camera and camera-radar 3D object detectors.
Authors: Yuhao Huang, Xin Yang, Lian Liu, Han Zhou, Ao Chang, Xinrui Zhou, Rusi Chen, Junxuan Yu, Jiongquan Chen, Chaoyu Chen, Sijing Liu, Haozhe Chi, Xindi Hu, Kejuan Yue, Lei Li, Vicente Grau, Deng-Ping Fan, Fajin Dong, Dong Ni
The Segment Anything Model (SAM) is the first foundation model for general image segmentation. It has achieved impressive results on various natural image segmentation tasks. However, medical image segmentation (MIS) is more challenging because of the complex modalities, fine anatomical structures, uncertain and complex object boundaries, and wide-range object scales. To fully validate SAM's performance on medical data, we collected and sorted 53 open-source datasets and built a large medical segmentation dataset with 18 modalities, 84 objects, 125 object-modality paired targets, 1050K 2D images, and 6033K masks. We comprehensively analyzed different models and strategies on the so-called COSMOS 1050K dataset. Our findings mainly include the following: 1) SAM showed remarkable performance in some specific objects but was unstable, imperfect, or even totally failed in other situations. 2) SAM with the large ViT-H showed better overall performance than that with the small ViT-B. 3) SAM performed better with manual hints, especially box, than the Everything mode. 4) SAM could help human annotation with high labeling quality and less time. 5) SAM was sensitive to the randomness in the center point and tight box prompts, and may suffer from a serious performance drop. 6) SAM performed better than interactive methods with one or a few points, but will be outpaced as the number of points increases. 7) SAM's performance correlated to different factors, including boundary complexity, intensity differences, etc. 8) Finetuning the SAM on specific medical tasks could improve its average DICE performance by 4.39% and 6.68% for ViT-B and ViT-H, respectively. We hope that this comprehensive report can help researchers explore the potential of SAM applications in MIS, and guide how to appropriately use and develop SAM.
Authors: Jonas Hein, Nicola Cavalcanti, Daniel Suter, Lukas Zingg, Fabio Carrillo, Lilian Calvet, Mazda Farshad, Marc Pollefeys, Nassir Navab, Philipp Fürnstahl
State-of-the-art research of traditional computer vision is increasingly leveraged in the surgical domain. A particular focus in computer-assisted surgery is to replace marker-based tracking systems for instrument localization with pure image-based 6DoF pose estimation using deep-learning methods. However, state-of-the-art single-view pose estimation methods do not yet meet the accuracy required for surgical navigation. In this context, we investigate the benefits of multi-view setups for highly accurate and occlusion-robust 6DoF pose estimation of surgical instruments and derive recommendations for an ideal camera system that addresses the challenges in the operating room.
The contributions of this work are threefold. First, we present a multi-camera capture setup consisting of static and head-mounted cameras, which allows us to study the performance of pose estimation methods under various camera configurations. Second, we publish a multi-view RGB-D video dataset of ex-vivo spine surgeries, captured in a surgical wet lab and a real operating theatre and including rich annotations for surgeon, instrument, and patient anatomy. Third, we evaluate three state-of-the-art single-view and multi-view methods for the task of 6DoF pose estimation of surgical instruments and analyze the influence of camera configurations, training data, and occlusions on the pose accuracy and generalization ability. The best method utilizes five cameras in a multi-view pose optimization and achieves an average position and orientation error of 1.01 mm and 0.89\deg for a surgical drill as well as 2.79 mm and 3.33\deg for a screwdriver under optimal conditions. Our results demonstrate that marker-less tracking of surgical instruments is becoming a feasible alternative to existing marker-based systems.
Authors: Xijun Wang, Aggelos K. Katsaggelos
Weakly-supervised temporal action localization aims to identify and localize the action instances in the untrimmed videos with only video-level action labels. When humans watch videos, we can adapt our abstract-level knowledge about actions in different video scenarios and detect whether some actions are occurring. In this paper, we mimic how humans do and bring a new perspective for locating and identifying multiple actions in a video. We propose a network named VQK-Net with a video-specific query-key attention modeling that learns a unique query for each action category of each input video. The learned queries not only contain the actions' knowledge features at the abstract level but also have the ability to fit this knowledge into the target video scenario, and they will be used to detect the presence of the corresponding action along the temporal dimension. To better learn these action category queries, we exploit not only the features of the current input video but also the correlation between different videos through a novel video-specific action category query learner worked with a query similarity loss. Finally, we conduct extensive experiments on three commonly used datasets (THUMOS14, ActivityNet1.2, and ActivityNet1.3) and achieve state-of-the-art performance.
Authors: Mitsuki Yoshida, Kanji Tanaka, Ryogo Yamamoto, Daiki Iwata
Semantic localization, i.e., robot self-localization with semantic image modality, is critical in recently emerging embodied AI applications (e.g., point-goal navigation, object-goal navigation, vision language navigation) and topological mapping applications (e.g., graph neural SLAM, ego-centric topological map). However, most existing works on semantic localization focus on passive vision tasks without viewpoint planning, or rely on additional rich modalities (e.g., depth measurements). Thus, the problem is largely unsolved. In this work, we explore a lightweight, entirely CPU-based, domain-adaptive semantic localization framework, called graph neural localizer. Our approach is inspired by two recently emerging technologies: (1) Scene graph, which combines the viewpoint- and appearance- invariance of local and global features; (2) Graph neural network, which enables direct learning/recognition of graph data (i.e., non-vector data). Specifically, a graph convolutional neural network is first trained as a scene graph classifier for passive vision, and then its knowledge is transferred to a reinforcement-learning planner for active vision. Experiments on two scenarios, self-supervised learning and unsupervised domain adaptation, using a photo-realistic Habitat simulator validate the effectiveness of the proposed method.
Authors: Yuhang Ling, Yuxi Li, Zhenye Gan, Jiangning Zhang, Mingmin Chi, Yabiao Wang
Audio-Visual Segmentation (AVS) is a challenging task, which aims to segment sounding objects in video frames by exploring audio signals. Generally AVS faces two key challenges: (1) Audio signals inherently exhibit a high degree of information density, as sounds produced by multiple objects are entangled within the same audio stream; (2) Objects of the same category tend to produce similar audio signals, making it difficult to distinguish between them and thus leading to unclear segmentation results. Toward this end, we propose TransAVS, the first Transformer-based end-to-end framework for AVS task. Specifically, TransAVS disentangles the audio stream as audio queries, which will interact with images and decode into segmentation masks with full transformer architectures. This scheme not only promotes comprehensive audio-image communication but also explicitly excavates instance cues encapsulated in the scene. Meanwhile, to encourage these audio queries to capture distinctive sounding objects instead of degrading to be homogeneous, we devise two self-supervised loss functions at both query and mask levels, allowing the model to capture distinctive features within similar audio data and achieve more precise segmentation. Our experiments demonstrate that TransAVS achieves state-of-the-art results on the AVSBench dataset, highlighting its effectiveness in bridging the gap between audio and visual modalities.
Authors: Linhui Xiao, Xiaoshan Yang, Fang Peng, Ming Yan, Yaowei Wang, Changsheng Xu
Visual Grounding (VG) is a crucial topic in the field of vision and language, which involves locating a specific region described by expressions within an image. To reduce the reliance on manually labeled data, unsupervised visual grounding have been developed to locate regions using pseudo-labels. However, the performance of existing unsupervised methods is highly dependent on the quality of pseudo-labels and these methods always encounter issues with limited diversity. In order to utilize vision and language pre-trained models to address the grounding problem, and reasonably take advantage of pseudo-labels, we propose CLIP-VG, a novel method that can conduct self-paced curriculum adapting of CLIP with pseudo-language labels. We propose a simple yet efficient end-to-end network architecture to realize the transfer of CLIP to the visual grounding. Based on the CLIP-based architecture, we further propose single-source and multi-source curriculum adapting algorithms, which can progressively find more reliable pseudo-labels to learn an optimal model, thereby achieving a balance between reliability and diversity for the pseudo-language labels. Our method outperforms the current state-of-the-art unsupervised method by a significant margin on RefCOCO/+/g datasets in both single-source and multi-source scenarios, with improvements ranging from 6.78$\%$ to 10.67$\%$ and 11.39$\%$ to 14.87$\%$, respectively. The results even outperform existing weakly supervised visual grounding methods. Furthermore, our method is also competitive in fully supervised setting. The code and models are available at https://github.com/linhuixiao/CLIP-VG.
Authors: Michal Yarom, Yonatan Bitton, Soravit Changpinyo, Roee Aharoni, Jonathan Herzig, Oran Lang, Eran Ofek, Idan Szpektor
Automatically determining whether a text and a corresponding image are semantically aligned is a significant challenge for vision-language models, with applications in generative text-to-image and image-to-text tasks. In this work, we study methods for automatic text-image alignment evaluation. We first introduce SeeTRUE: a comprehensive evaluation set, spanning multiple datasets from both text-to-image and image-to-text generation tasks, with human judgements for whether a given text-image pair is semantically aligned. We then describe two automatic methods to determine alignment: the first involving a pipeline based on question generation and visual question answering models, and the second employing an end-to-end classification approach by finetuning multimodal pretrained models. Both methods surpass prior approaches in various text-image alignment tasks, with significant improvements in challenging cases that involve complex composition or unnatural images. Finally, we demonstrate how our approaches can localize specific misalignments between an image and a given text, and how they can be used to automatically re-rank candidates in text-to-image generation.
Authors: Dong Wei, Xiaoning Sun, Huaijiang Sun, Bin Li, Shengxiang Hu, Weiqing Li, Jianfeng Lu
The emergence of text-driven motion synthesis technique provides animators with great potential to create efficiently. However, in most cases, textual expressions only contain general and qualitative motion descriptions, while lack fine depiction and sufficient intensity, leading to the synthesized motions that either (a) semantically compliant but uncontrollable over specific pose details, or (b) even deviates from the provided descriptions, bringing animators with undesired cases. In this paper, we propose DiffKFC, a conditional diffusion model for text-driven motion synthesis with KeyFrames Collaborated, enabling realistic generation with collaborative and efficient dual-level control: coarse guidance at semantic level, with only few keyframes for direct and fine-grained depiction down to body posture level. Unlike existing inference-editing diffusion models that incorporate conditions without training, our conditional diffusion model is explicitly trained and can fully exploit correlations among texts, keyframes and the diffused target frames. To preserve the control capability of discrete and sparse keyframes, we customize dilated mask attention modules where only partial valid tokens participate in local-to-global attention, indicated by the dilated keyframe mask. Additionally, we develop a simple yet effective smoothness prior, which steers the generated frames towards seamless keyframe transitions at inference. Extensive experiments show that our model not only achieves state-of-the-art performance in terms of semantic fidelity, but more importantly, is able to satisfy animator requirements through fine-grained guidance without tedious labor.
Authors: Jingyuan Sun, Mingxiao Li, Zijiao Chen, Yunhao Zhang, Shaonan Wang, Marie-Francine Moens
Decoding visual stimuli from neural responses recorded by functional Magnetic Resonance Imaging (fMRI) presents an intriguing intersection between cognitive neuroscience and machine learning, promising advancements in understanding human visual perception and building non-invasive brain-machine interfaces. However, the task is challenging due to the noisy nature of fMRI signals and the intricate pattern of brain visual representations. To mitigate these challenges, we introduce a two-phase fMRI representation learning framework. The first phase pre-trains an fMRI feature learner with a proposed Double-contrastive Mask Auto-encoder to learn denoised representations. The second phase tunes the feature learner to attend to neural activation patterns most informative for visual reconstruction with guidance from an image auto-encoder. The optimized fMRI feature learner then conditions a latent diffusion model to reconstruct image stimuli from brain activities. Experimental results demonstrate our model's superiority in generating high-resolution and semantically accurate images, substantially exceeding previous state-of-the-art methods by 39.34% in the 50-way-top-1 semantic classification accuracy. Our research invites further exploration of the decoding task's potential and contributes to the development of non-invasive brain-machine interfaces.
Authors: Jiacheng Chen, Ruizhi Deng, Yasutaka Furukawa
This paper presents PolyDiffuse, a novel structured reconstruction algorithm that transforms visual sensor data into polygonal shapes with Diffusion Models (DM), an emerging machinery amid exploding generative AI, while formulating reconstruction as a generation process conditioned on sensor data. The task of structured reconstruction poses two fundamental challenges to DM: 1) A structured geometry is a ``set'' (e.g., a set of polygons for a floorplan geometry), where a sample of $N$ elements has $N!$ different but equivalent representations, making the denoising highly ambiguous; and 2) A ``reconstruction'' task has a single solution, where an initial noise needs to be chosen carefully, while any initial noise works for a generation task. Our technical contribution is the introduction of a Guided Set Diffusion Model where 1) the forward diffusion process learns guidance networks to control noise injection so that one representation of a sample remains distinct from its other permutation variants, thus resolving denoising ambiguity; and 2) the reverse denoising process reconstructs polygonal shapes, initialized and directed by the guidance networks, as a conditional generation process subject to the sensor data. We have evaluated our approach for reconstructing two types of polygonal shapes: floorplan as a set of polygons and HD map for autonomous cars as a set of polylines. Through extensive experiments on standard benchmarks, we demonstrate that PolyDiffuse significantly advances the current state of the art and enables broader practical applications.
Authors: Alexandre Lacoste, Nils Lehmann, Pau Rodriguez, Evan David Sherwin, Hannah Kerner, Björn Lütjens, Jeremy Andrew Irvin, David Dao, Hamed Alemohammad, Alexandre Drouin, Mehmet Gunturkun, Gabriel Huang, David Vazquez, Dava Newman, Yoshua Bengio, Stefano Ermon, Xiao Xiang Zhu
Recent progress in self-supervision has shown that pre-training large neural networks on vast amounts of unsupervised data can lead to substantial increases in generalization to downstream tasks. Such models, recently coined foundation models, have been transformational to the field of natural language processing. Variants have also been proposed for image data, but their applicability to remote sensing tasks is limited. To stimulate the development of foundation models for Earth monitoring, we propose a benchmark comprised of six classification and six segmentation tasks, which were carefully curated and adapted to be both relevant to the field and well-suited for model evaluation. We accompany this benchmark with a robust methodology for evaluating models and reporting aggregated results to enable a reliable assessment of progress. Finally, we report results for 20 baselines to gain information about the performance of existing models. We believe that this benchmark will be a driver of progress across a variety of Earth monitoring tasks.
Authors: Bipasha Sen, Gaurav Singh, Aditya Agarwal, Rohith Agaram, K Madhava Krishna, Srinath Sridhar
Neural Radiance Fields (NeRF) have become an increasingly popular representation to capture high-quality appearance and shape of scenes and objects. However, learning generalizable NeRF priors over categories of scenes or objects has been challenging due to the high dimensionality of network weight space. To address the limitations of existing work on generalization, multi-view consistency and to improve quality, we propose HyP-NeRF, a latent conditioning method for learning generalizable category-level NeRF priors using hypernetworks. Rather than using hypernetworks to estimate only the weights of a NeRF, we estimate both the weights and the multi-resolution hash encodings resulting in significant quality gains. To improve quality even further, we incorporate a denoise and finetune strategy that denoises images rendered from NeRFs estimated by the hypernetwork and finetunes it while retaining multiview consistency. These improvements enable us to use HyP-NeRF as a generalizable prior for multiple downstream tasks including NeRF reconstruction from single-view or cluttered scenes and text-to-NeRF. We provide qualitative comparisons and evaluate HyP-NeRF on three tasks: generalization, compression, and retrieval, demonstrating our state-of-the-art results.
Authors: Nikolaos Louloudakis, Perry Gibson, José Cano, Ajitha Rajan
When deploying Deep Neural Networks (DNNs), developers often convert models from one deep learning framework to another (e.g., TensorFlow to PyTorch). However, this process is error-prone and can impact target model accuracy. To identify the extent of such impact, we perform and briefly present a differential analysis against three DNNs widely used for image recognition (MobileNetV2, ResNet101, and InceptionV3) converted across four well-known deep learning frameworks (PyTorch, Keras, TensorFlow (TF), and TFLite), which revealed numerous model crashes and output label discrepancies of up to 72%. To mitigate such errors, we present a novel approach towards fault localization and repair of buggy deep learning framework conversions, focusing on pre-trained image recognition models. Our technique consists of four stages of analysis: 1) conversion tools, 2) model parameters, 3) model hyperparameters, and 4) graph representation. In addition, we propose various strategies towards fault repair of the faults detected. We implement our technique on top of the Apache TVM deep learning compiler, and we test it by conducting a preliminary fault localization analysis for the conversion of InceptionV3 from TF to TFLite. Our approach detected a fault in a common DNN converter tool, which introduced precision errors in weights, reducing model accuracy. After our fault localization, we repaired the issue, reducing our conversion error to zero.
Authors: Nikolaos Louloudakis, Perry Gibson, José Cano, Ajitha Rajan
Image recognition tasks typically use deep learning and require enormous processing power, thus relying on hardware accelerators like GPUs and TPUs for fast, timely processing. Failure in real-time image recognition tasks can occur due to sub-optimal mapping on hardware accelerators during model deployment, which may lead to timing uncertainty and erroneous behavior. Mapping on hardware accelerators is done using multiple software components like deep learning frameworks, compilers, and device libraries, that we refer to as the computational environment. Owing to the increased use of image recognition tasks in safety-critical applications like autonomous driving and medical imaging, it is imperative to assess their robustness to changes in the computational environment, as the impact of parameters like deep learning frameworks, compiler optimizations, and hardware devices on model performance and correctness is not yet well understood.
In this paper we present a differential testing framework, DeltaNN, that allows us to assess the impact of different computational environment parameters on the performance of image recognition models during deployment, post training. DeltaNN generates different implementations of a given image recognition model for variations in environment parameters, namely, deep learning frameworks, compiler optimizations and hardware devices and analyzes differences in model performance as a result. Using DeltaNN, we conduct an empirical study of robustness analysis of three popular image recognition models using the ImageNet dataset. We report the impact in terms of misclassifications and inference time differences across different settings. In total, we observed up to 72% output label differences across deep learning frameworks, and up to 81% unexpected performance degradation in terms of inference time, when applying compiler optimizations.
Authors: Xiaoyang Hao (1 and 2), Han Li (1), Jun Cheng (2), Lei Wang (2) ((1) Southern University of Science and Technology, (2) Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences)
Existing methods for 3D human mesh recovery always directly estimate SMPL parameters, which involve both joint rotations and shape parameters. However, these methods present rotation semantic ambiguity, rotation error accumulation, and shape estimation overfitting, which also leads to errors in the estimated pose. Additionally, these methods have not efficiently leveraged the advancements in another hot topic, human pose estimation. To address these issues, we propose a novel approach, Decomposition of 3D Rotation and Lift from 2D Joint to 3D mesh (D3L). We disentangle 3D joint rotation into bone direction and bone twist direction so that the human mesh recovery task is broken down into estimation of pose, twist, and shape, which can be handled independently. Then we design a 2D-to-3D lifting network for estimating twist direction and 3D joint position from 2D joint position sequences and introduce a nonlinear optimization method for fitting shape parameters and bone directions. Our approach can leverage human pose estimation methods, and avoid pose errors introduced by shape estimation overfitting. We conduct experiments on the Human3.6M dataset and demonstrate improved performance compared to existing methods by a large margin.
Authors: Jiyao Zhang, Mingdong Wu, Hao Dong
Object pose estimation plays a vital role in embodied AI and computer vision, enabling intelligent agents to comprehend and interact with their surroundings. Despite the practicality of category-level pose estimation, current approaches encounter challenges with partially observed point clouds, known as the multihypothesis issue. In this study, we propose a novel solution by reframing categorylevel object pose estimation as conditional generative modeling, departing from traditional point-to-point regression. Leveraging score-based diffusion models, we estimate object poses by sampling candidates from the diffusion model and aggregating them through a two-step process: filtering out outliers via likelihood estimation and subsequently mean-pooling the remaining candidates. To avoid the costly integration process when estimating the likelihood, we introduce an alternative method that trains an energy-based model from the original score-based model, enabling end-to-end likelihood estimation. Our approach achieves state-of-the-art performance on the REAL275 dataset, surpassing 50% and 60% on strict 5d2cm and 5d5cm metrics, respectively. Furthermore, our method demonstrates strong generalizability to novel categories sharing similar symmetric properties without fine-tuning and can readily adapt to object pose tracking tasks, yielding comparable results to the current state-of-the-art baselines.
Authors: Marcos V. Conde, Javier Vazquez-Corral, Michael S. Brown, Radu Timofte
3D lookup tables (3D LUTs) are a key component for image enhancement. Modern image signal processors (ISPs) have dedicated support for these as part of the camera rendering pipeline. Cameras typically provide multiple options for picture styles, where each style is usually obtained by applying a unique handcrafted 3D LUT. Current approaches for learning and applying 3D LUTs are notably fast, yet not so memory-efficient, as storing multiple 3D LUTs is required. For this reason and other implementation limitations, their use on mobile devices is less popular. In this work, we propose a Neural Implicit LUT (NILUT), an implicitly defined continuous 3D color transformation parameterized by a neural network. We show that NILUTs are capable of accurately emulating real 3D LUTs. Moreover, a NILUT can be extended to incorporate multiple styles into a single network with the ability to blend styles implicitly. Our novel approach is memory-efficient, controllable and can complement previous methods, including learned ISPs. Code, models and dataset available at: https://github.com/mv-lab/nilut
Authors: Satoshi Tsutsui, Winnie Pang, Bihan Wen
The examination of blood samples at a microscopic level plays a fundamental role in clinical diagnostics, influencing a wide range of medical conditions. For instance, an in-depth study of White Blood Cells (WBCs), a crucial component of our blood, is essential for diagnosing blood-related diseases such as leukemia and anemia. While multiple datasets containing WBC images have been proposed, they mostly focus on cell categorization, often lacking the necessary morphological details to explain such categorizations, despite the importance of explainable artificial intelligence (XAI) in medical domains. This paper seeks to address this limitation by introducing comprehensive annotations for WBC images. Through collaboration with pathologists, a thorough literature review, and manual inspection of microscopic images, we have identified 11 morphological attributes associated with the cell and its components (nucleus, cytoplasm, and granules). We then annotated ten thousand WBC images with these attributes. Moreover, we conduct experiments to predict these attributes from images, providing insights beyond basic WBC classification. As the first public dataset to offer such extensive annotations, we also illustrate specific applications that can benefit from our attribute annotations. Overall, our dataset paves the way for interpreting WBC recognition models, further advancing XAI in the fields of pathology and hematology.
Authors: Tianxiang Ma, Kang Zhao, Jianxin Sun, Yingya Zhang, Jing Dong
Efficiently generating a freestyle 3D portrait with high quality and 3D-consistency is a promising yet challenging task. The portrait styles generated by most existing methods are usually restricted by their 3D generators, which are learned in specific facial datasets, such as FFHQ. To get the diverse 3D portraits, one can build a large-scale multi-style database to retrain a 3D-aware generator, or use a off-the-shelf tool to do the style translation. However, the former is time-consuming due to data collection and training process, the latter may destroy the multi-view consistency. To tackle this problem, we propose a novel text-driven 3D-aware portrait synthesis framework that can generate out-of-distribution portrait styles. Specifically, for a given portrait style prompt, we first composite two generative priors, a 3D-aware GAN generator and a text-guided image editor, to quickly construct a few-shot stylized portrait set. Then we map the special style domain of this set to our proposed 3D latent feature generator and obtain a 3D representation containing the given style information. Finally we use a pre-trained 3D renderer to generate view-consistent stylized portraits from the 3D representation. Extensive experimental results show that our method is capable of synthesizing high-quality 3D portraits with specified styles in a few minutes, outperforming the state-of-the-art.
Authors: Shitao Tang, Fuyang Zhang, Jiacheng Chen, Peng Wang, Yasutaka Furukawa
This paper introduces MVDiffusion, a simple yet effective method for generating consistent multi-view images from text prompts given pixel-to-pixel correspondences (e.g., perspective crops from a panorama or multi-view images given depth maps and poses). Unlike prior methods that rely on iterative image warping and inpainting, MVDiffusion simultaneously generates all images with a global awareness, effectively addressing the prevalent error accumulation issue. At its core, MVDiffusion processes perspective images in parallel with a pre-trained text-to-image diffusion model, while integrating novel correspondence-aware attention layers to facilitate cross-view interactions. For panorama generation, while only trained with 10k panoramas, MVDiffusion is able to generate high-resolution photorealistic images for arbitrary texts or extrapolate one perspective image to a 360-degree view. For multi-view depth-to-image generation, MVDiffusion demonstrates state-of-the-art performance for texturing a scene mesh.
Authors: Yuxiang Zhang, Hongwen Zhang, Liangxiao Hu, Jiajun Zhang, Hongwei Yi, Shengping Zhang, Yebin Liu
Learning-based approaches to monocular motion capture have recently shown promising results by learning to regress in a data-driven manner. However, due to the challenges in data collection and network designs, it remains challenging for existing solutions to achieve real-time full-body capture while being accurate in world space. In this work, we introduce ProxyCap, a human-centric proxy-to-motion learning scheme to learn world-space motions from a proxy dataset of 2D skeleton sequences and 3D rotational motions. Such proxy data enables us to build a learning-based network with accurate world-space supervision while also mitigating the generalization issues. For more accurate and physically plausible predictions in world space, our network is designed to learn human motions from a human-centric perspective, which enables the understanding of the same motion captured with different camera trajectories. Moreover, a contact-aware neural motion descent module is proposed in our network so that it can be aware of foot-ground contact and motion misalignment with the proxy observations. With the proposed learning-based solution, we demonstrate the first real-time monocular full-body capture system with plausible foot-ground contact in world space even using hand-held moving cameras. Our project page is https://zhangyux15.github.io/ProxyCapV2.
Authors: Tom Monnier, Jake Austin, Angjoo Kanazawa, Alexei A. Efros, Mathieu Aubry
Given a set of calibrated images of a scene, we present an approach that produces a simple, compact, and actionable 3D world representation by means of 3D primitives. While many approaches focus on recovering high-fidelity 3D scenes, we focus on parsing a scene into mid-level 3D representations made of a small set of textured primitives. Such representations are interpretable, easy to manipulate and suited for physics-based simulations. Moreover, unlike existing primitive decomposition methods that rely on 3D input data, our approach operates directly on images through differentiable rendering. Specifically, we model primitives as textured superquadric meshes and optimize their parameters from scratch with an image rendering loss. We highlight the importance of modeling transparency for each primitive, which is critical for optimization and also enables handling varying numbers of primitives. We show that the resulting textured primitives faithfully reconstruct the input images and accurately model the visible 3D points, while providing amodal shape completions of unseen object regions. We compare our approach to the state of the art on diverse scenes from DTU, and demonstrate its robustness on real-life captures from BlendedMVS and Nerfstudio. We also showcase how our results can be used to effortlessly edit a scene or perform physical simulations. Code and video results are available at https://www.tmonnier.com/DBW .
Authors: Lana Touma, Mohammad Al Horani, Manar Tailouni, Anas Dahabiah, Khloud Al Jallad
Automatic Deception Detection has been a hot research topic for a long time, using machine learning and deep learning to automatically detect deception, brings new light to this old field. In this paper, we proposed a voting-based method for automatic deception detection from videos using audio, visual and lexical features. Experiments were done on two datasets, the Real-life trial dataset by Michigan University and the Miami University deception detection dataset. Video samples were split into frames of images, audio, and manuscripts. Our Voting-based Multimodal proposed solution consists of three models. The first model is CNN for detecting deception from images, the second model is Support Vector Machine (SVM) on Mel spectrograms for detecting deception from audio and the third model is Word2Vec on Support Vector Machine (SVM) for detecting deception from manuscripts. Our proposed solution outperforms state of the art. Best results achieved on images, audio and text were 97%, 96%, 92% respectively on Real-Life Trial Dataset, and 97%, 82%, 73% on video, audio and text respectively on Miami University Deception Detection.
Authors: Nikita Karaev, Ignacio Rocco, Benjamin Graham, Natalia Neverova, Andrea Vedaldi, Christian Rupprecht
We introduce CoTracker, a transformer-based model that tracks dense points in a frame jointly across a video sequence. This differs from most existing state-of-the-art approaches that track points independently, ignoring their correlation. We show that joint tracking results in a significantly higher tracking accuracy and robustness. We also provide several technical innovations, including the concept of virtual tracks, which allows CoTracker to track 70k points jointly and simultaneously. Furthermore, CoTracker operates causally on short windows (hence, it is suitable for online tasks), but is trained by unrolling the windows across longer video sequences, which enables and significantly improves long-term tracking. We demonstrate qualitatively impressive tracking results, where points can be tracked for a long time even when they are occluded or leave the field of view. Quantitatively, CoTracker outperforms all recent trackers on standard benchmarks, often by a substantial margin.
Authors: Ke Cao, Ruiping Liu, Ze Wang, Kunyu Peng, Jiaming Zhang, Junwei Zheng, Zhifeng Teng, Kailun Yang, Rainer Stiefelhagen
The mobile robot relies on SLAM (Simultaneous Localization and Mapping) to provide autonomous navigation and task execution in complex and unknown environments. However, it is hard to develop a dedicated algorithm for mobile robots due to dynamic and challenging situations, such as poor lighting conditions and motion blur. To tackle this issue, we propose a tightly-coupled LiDAR-visual SLAM based on geometric features, which includes two sub-systems (LiDAR and monocular visual SLAM) and a fusion framework. The fusion framework associates the depth and semantics of the multi-modal geometric features to complement the visual line landmarks and to add direction optimization in Bundle Adjustment (BA). This further constrains visual odometry. On the other hand, the entire line segment detected by the visual subsystem overcomes the limitation of the LiDAR subsystem, which can only perform the local calculation for geometric features. It adjusts the direction of linear feature points and filters out outliers, leading to a higher accurate odometry system. Finally, we employ a module to detect the subsystem's operation, providing the LiDAR subsystem's output as a complementary trajectory to our system while visual subsystem tracking fails. The evaluation results on the public dataset M2DGR, gathered from ground robots across various indoor and outdoor scenarios, show that our system achieves more accurate and robust pose estimation compared to current state-of-the-art multi-modal methods.
Authors: Zikai Zhou, Huanran Chen
The success of deep learning is inseparable from normalization layers. Researchers have proposed various normalization functions, and each of them has both advantages and disadvantages. In response, efforts have been made to design a unified normalization function that combines all normalization procedures and mitigates their weaknesses. We also proposed a new normalization function called Adaptive Fusion Normalization. Through experiments, we demonstrate AFN outperforms the previous normalization techniques in domain generalization and image classification tasks.
Authors: Xin Jin, Jia-Wen Xiao, Ling-Hao Han, Chunle Guo, Xialei Liu, Chongyi Li, Ming-Ming Cheng
Explicit calibration-based methods have dominated RAW image denoising under extremely low-light environments. However, these methods are impeded by several critical limitations: a) the explicit calibration process is both labor- and time-intensive, b) challenge exists in transferring denoisers across different camera models, and c) the disparity between synthetic and real noise is exacerbated by digital gain. To address these issues, we introduce a groundbreaking pipeline named Lighting Every Darkness (LED), which is effective regardless of the digital gain or the camera sensor. LED eliminates the need for explicit noise model calibration, instead utilizing an implicit fine-tuning process that allows quick deployment and requires minimal data. Structural modifications are also included to reduce the discrepancy between synthetic and real noise without extra computational demands. Our method surpasses existing methods in various camera models, including new ones not in public datasets, with just a few pairs per digital gain and only 0.5% of the typical iterations. Furthermore, LED also allows researchers to focus more on deep learning advancements while still utilizing sensor engineering benefits. Code and related materials can be found in https://srameo.github.io/projects/led-iccv23/ .
Authors: Zixuan Pan, Jianxu Chen, Yiyu Shi
Denoising diffusion probabilistic models have recently demonstrated state-of-the-art generative performance and have been used as strong pixel-level representation learners. This paper decomposes the interrelation between the generative capability and representation learning ability inherent in diffusion models. We present the masked diffusion model (MDM), a scalable self-supervised representation learner for semantic segmentation, substituting the conventional additive Gaussian noise of traditional diffusion with a masking mechanism. Our proposed approach convincingly surpasses prior benchmarks, demonstrating remarkable advancements in both medical and natural image semantic segmentation tasks, particularly in few-shot scenarios.
Authors: Yonatan Bitton, Hritik Bansal, Jack Hessel, Rulin Shao, Wanrong Zhu, Anas Awadalla, Josh Gardner, Rohan Taori, Ludwig Schmidt
We introduce VisIT-Bench (Visual InsTruction Benchmark), a benchmark for evaluation of instruction-following vision-language models for real-world use. Our starting point is curating 70 'instruction families' that we envision instruction tuned vision-language models should be able to address. Extending beyond evaluations like VQAv2 and COCO, tasks range from basic recognition to game playing and creative generation. Following curation, our dataset comprises 592 test queries, each with a human-authored instruction-conditioned caption. These descriptions surface instruction-specific factors, e.g., for an instruction asking about the accessibility of a storefront for wheelchair users, the instruction-conditioned caption describes ramps/potential obstacles. These descriptions enable 1) collecting human-verified reference outputs for each instance; and 2) automatic evaluation of candidate multimodal generations using a text-only LLM, aligning with human judgment. We quantify quality gaps between models and references using both human and automatic evaluations; e.g., the top-performing instruction-following model wins against the GPT-4 reference in just 27% of the comparison. VisIT-Bench is dynamic to participate, practitioners simply submit their model's response on the project website; Data, code and leaderboard is available at visit-bench.github.io.
Authors: Haojie Huang, Dian Wang, Arsh Tangri, Robin Walters, Robert Platt
Robotic pick and place tasks are symmetric under translations and rotations of both the object to be picked and the desired place pose. For example, if the pick object is rotated or translated, then the optimal pick action should also rotate or translate. The same is true for the place pose; if the desired place pose changes, then the place action should also transform accordingly. A recently proposed pick and place framework known as Transporter Net captures some of these symmetries, but not all. This paper analytically studies the symmetries present in planar robotic pick and place and proposes a method of incorporating equivariant neural models into Transporter Net in a way that captures all symmetries. The new model, which we call Equivariant Transporter Net, is equivariant to both pick and place symmetries and can immediately generalize pick and place knowledge to different pick and place poses. We evaluate the new model empirically and show that it is much more sample efficient than the non-symmetric version, resulting in a system that can imitate demonstrated pick and place behavior using very few human demonstrations on a variety of imitation learning tasks.
Authors: Song Tang, Chuang Li, Pu Zhang, RongNian Tang
Integrating CNNs and RNNs to capture spatiotemporal dependencies is a prevalent strategy for spatiotemporal prediction tasks. However, the property of CNNs to learn local spatial information decreases their efficiency in capturing spatiotemporal dependencies, thereby limiting their prediction accuracy. In this paper, we propose a new recurrent cell, SwinLSTM, which integrates Swin Transformer blocks and the simplified LSTM, an extension that replaces the convolutional structure in ConvLSTM with the self-attention mechanism. Furthermore, we construct a network with SwinLSTM cell as the core for spatiotemporal prediction. Without using unique tricks, SwinLSTM outperforms state-of-the-art methods on Moving MNIST, Human3.6m, TaxiBJ, and KTH datasets. In particular, it exhibits a significant improvement in prediction accuracy compared to ConvLSTM. Our competitive experimental results demonstrate that learning global spatial dependencies is more advantageous for models to capture spatiotemporal dependencies. We hope that SwinLSTM can serve as a solid baseline to promote the advancement of spatiotemporal prediction accuracy. The codes are publicly available at https://github.com/SongTang-x/SwinLSTM.
Authors: Qingwen Bu, Sungrae Park, Minsoo Khang, Yichuan Cheng
Existing techniques for text detection can be broadly classified into two primary groups: segmentation-based and regression-based methods. Segmentation models offer enhanced robustness to font variations but require intricate post-processing, leading to high computational overhead. Regression-based methods undertake instance-aware prediction but face limitations in robustness and data efficiency due to their reliance on high-level representations. In our academic pursuit, we propose SRFormer, a unified DETR-based model with amalgamated Segmentation and Regression, aiming at the synergistic harnessing of the inherent robustness in segmentation representations, along with the straightforward post-processing of instance-level regression. Our empirical analysis indicates that favorable segmentation predictions can be obtained at the initial decoder layers. In light of this, we constrain the incorporation of segmentation branches to the first few decoder layers and employ progressive regression refinement in subsequent layers, achieving performance gains while minimizing computational load from the mask.Furthermore, we propose a Mask-informed Query Enhancement module. We take the segmentation result as a natural soft-ROI to pool and extract robust pixel representations, which are then employed to enhance and diversify instance queries. Extensive experimentation across multiple benchmarks has yielded compelling findings, highlighting our method's exceptional robustness, superior training and data efficiency, as well as its state-of-the-art performance. Our code is available at https://github.com/retsuh-bqw/SRFormer-Text-Det.
Authors: Christopher Hahne, Michel Hayoz, Raphael Sznitman
Time of Flight (ToF) is a prevalent depth sensing technology in the fields of robotics, medical imaging, and non-destructive testing. Yet, ToF sensing faces challenges from complex ambient conditions making an inverse modelling from the sparse temporal information intractable. This paper highlights the potential of modern super-resolution techniques to learn varying surroundings for a reliable and accurate ToF detection. Unlike existing models, we tailor an architecture for sub-sample precise semi-global signal localization by combining super-resolution with an efficient residual contraction block to balance between fine signal details and large scale contextual information. We consolidate research on ToF by conducting a benchmark comparison against six state-of-the-art methods for which we employ two publicly available datasets. This includes the release of our SToF-Chirp dataset captured by an airborne ultrasound transducer. Results showcase the superior performance of our proposed StofNet in terms of precision, reliability and model complexity. Our code is available at https://github.com/hahnec/stofnet.
Authors: Baowen Zhang, Jiahe Li, Xiaoming Deng, Yinda Zhang, Cuixia Ma, Hongan Wang
Learning 3D shape representation with dense correspondence for deformable objects is a fundamental problem in computer vision. Existing approaches often need additional annotations of specific semantic domain, e.g., skeleton poses for human bodies or animals, which require extra annotation effort and suffer from error accumulation, and they are limited to specific domain. In this paper, we propose a novel self-supervised approach to learn neural implicit shape representation for deformable objects, which can represent shapes with a template shape and dense correspondence in 3D. Our method does not require the priors of skeleton and skinning weight, and only requires a collection of shapes represented in signed distance fields. To handle the large deformation, we constrain the learned template shape in the same latent space with the training shapes, design a new formulation of local rigid constraint that enforces rigid transformation in local region and addresses local reflection issue, and present a new hierarchical rigid constraint to reduce the ambiguity due to the joint learning of template shape and correspondences. Extensive experiments show that our model can represent shapes with large deformations. We also show that our shape representation can support two typical applications, such as texture transfer and shape editing, with competitive performance. The code and models are available at https://iscas3dv.github.io/deformshape
Authors: Sheng Zhang, Muzammal Naseer, Guangyi Chen, Zhiqiang Shen, Salman Khan, Kun Zhang, Fahad Khan
Large-scale pre-trained Vision Language Models (VLMs) have proven effective for zero-shot classification. Despite the success, most traditional VLMs-based methods are restricted by the assumption of partial source supervision or ideal vocabularies, which rarely satisfy the open-world scenario. In this paper, we aim at a more challenging setting, Realistic Zero-Shot Classification, which assumes no annotation but instead a broad vocabulary. To address this challenge, we propose the Self Structural Semantic Alignment (S^3A) framework, which extracts the structural semantic information from unlabeled data while simultaneously self-learning. Our S^3A framework adopts a unique Cluster-Vote-Prompt-Realign (CVPR) algorithm, which iteratively groups unlabeled data to derive structural semantics for pseudo-supervision. Our CVPR process includes iterative clustering on images, voting within each cluster to identify initial class candidates from the vocabulary, generating discriminative prompts with large language models to discern confusing candidates, and realigning images and the vocabulary as structural semantic alignment. Finally, we propose to self-learn the CLIP image encoder with both individual and structural semantic alignment through a teacher-student learning strategy. Our comprehensive experiments across various generic and fine-grained benchmarks demonstrate that the S^3A method offers substantial improvements over existing VLMs-based approaches, achieving a more than 15% accuracy improvement over CLIP on average. Our codes, models, and prompts are publicly released at https://github.com/sheng-eatamath/S3A.
Authors: Suxing Liu
Despite the remarkable results of deep learning in breast cancer histopathology image classification, challenges such as data imbalance and interpretability still exist and require cross-domain knowledge and collaboration among medical experts. This study proposes a breast cancer classification method using a dual-activated lightweight attention ResNet50 model, effectively addressing data imbalance and interpretability challenges. The model fuses a pre-trained deep ResNet50 and a lightweight attention mechanism to accomplish classification by embedding an attention module in layer 4 of ResNet50 and adding two fully connected layers. The fully connected network design employs LeakyReLU and ReLU activation functions.
The model outperforms SEResNet50, DensNet121, VGG16, VGG16Inception, ViT, Swin- Transformer, Dinov2_Vitb14, and ResNet50 models regarding precision, accuracy, recall, F1 score, and GMean, especially in the application performance on the BreakHis dataset. In particular, the model demonstrates significant robustness and broad applicability when dealing with the unbalanced breast cancer dataset. The model has been evaluated on histopathology images at magnification factors of 40X, 100X, 200X, and 400X, achieving accuracies of 98.5%, 98.7%, 97.9%, and 94.3%, respectively. The study comprehensively assessed the model's performance. In the later stages of training, the validated losses and accuracies change minimally, showing that the model avoids overfitting and exhibits good generalization ability. This model exhibited the fastest convergence in all laboratory experiments, even though its parameters are not the smallest. This highlights the model's efficacy as a lightweight attention framework, showcasing its efficiency in achieving rapid convergence without compromising performance.
Authors: Shuting He, Henghui Ding, Chang Liu, Xudong Jiang
The objective of Classic Referring Expression Comprehension (REC) is to produce a bounding box corresponding to the object mentioned in a given textual description. Commonly, existing datasets and techniques in classic REC are tailored for expressions that pertain to a single target, meaning a sole expression is linked to one specific object. Expressions that refer to multiple targets or involve no specific target have not been taken into account. This constraint hinders the practical applicability of REC. This study introduces a new benchmark termed as Generalized Referring Expression Comprehension (GREC). This benchmark extends the classic REC by permitting expressions to describe any number of target objects. To achieve this goal, we have built the first large-scale GREC dataset named gRefCOCO. This dataset encompasses a range of expressions: those referring to multiple targets, expressions with no specific target, and the single-target expressions. The design of GREC and gRefCOCO ensures smooth compatibility with classic REC. The proposed gRefCOCO dataset, a GREC method implementation code, and GREC evaluation code are available at https://github.com/henghuiding/gRefCOCO.
Authors: Yuhong He, Long Peng, Lu Wang, Jun Cheng
Since rain streaks show a variety of shapes and directions, learning the degradation representation is extremely challenging for single image deraining. Existing methods are mainly targeted at designing complicated modules to implicitly learn latent degradation representation from coupled rainy images. This way, it is hard to decouple the content-independent degradation representation due to the lack of explicit constraint, resulting in over- or under-enhancement problems. To tackle this issue, we propose a novel Latent Degradation Representation Constraint Network (LDRCNet) that consists of Direction-Aware Encoder (DAEncoder), UNet Deraining Network, and Multi-Scale Interaction Block (MSIBlock). Specifically, the DAEncoder is proposed to adaptively extract latent degradation representation by using the deformable convolutions to exploit the direction consistency of rain streaks. Next, a constraint loss is introduced to explicitly constraint the degradation representation learning during training. Last, we propose an MSIBlock to fuse with the learned degradation representation and decoder features of the deraining network for adaptive information interaction, which enables the deraining network to remove various complicated rainy patterns and reconstruct image details. Experimental results on synthetic and real datasets demonstrate that our method achieves new state-of-the-art performance.
Authors: Sheng Ye, Yubin Hu, Matthieu Lin, Yu-Hui Wen, Wang Zhao, Yong-Jin Liu, Wenping Wang
The reconstruction of indoor scenes from multi-view RGB images is challenging due to the coexistence of flat and texture-less regions alongside delicate and fine-grained regions. Recent methods leverage neural radiance fields aided by predicted surface normal priors to recover the scene geometry. These methods excel in producing complete and smooth results for floor and wall areas. However, they struggle to capture complex surfaces with high-frequency structures due to the inadequate neural representation and the inaccurately predicted normal priors. This work aims to reconstruct high-fidelity surfaces with fine-grained details by addressing the above limitations. To improve the capacity of the implicit representation, we propose a hybrid architecture to represent low-frequency and high-frequency regions separately. To enhance the normal priors, we introduce a simple yet effective image sharpening and denoising technique, coupled with a network that estimates the pixel-wise uncertainty of the predicted surface normal vectors. Identifying such uncertainty can prevent our model from being misled by unreliable surface normal supervisions that hinder the accurate reconstruction of intricate geometries. Experiments on the benchmark datasets show that our method outperforms existing methods in terms of reconstruction quality. Furthermore, the proposed method also generalizes well to real-world indoor scenarios captured by our hand-held mobile phones. Our code is publicly available at: https://github.com/yec22/Fine-Grained-Indoor-Recon.
Authors: Qiao Yang, Yu Zhang, Jian Zhang, Zijing Zhao, Shunli Zhang, Jinqiao Wang, Junzhe Chen
Most existing learning-based infrared and visible image fusion (IVIF) methods exhibit massive redundant information in the fusion images, i.e., yielding edge-blurring effect or unrecognizable for object detectors. To alleviate these issues, we propose a semantic structure-preserving approach for IVIF, namely SSPFusion. At first, we design a Structural Feature Extractor (SFE) to extract the structural features of infrared and visible images. Then, we introduce a multi-scale Structure-Preserving Fusion (SPF) module to fuse the structural features of infrared and visible images, while maintaining the consistency of semantic structures between the fusion and source images. Owing to these two effective modules, our method is able to generate high-quality fusion images from pairs of infrared and visible images, which can boost the performance of downstream computer-vision tasks. Experimental results on three benchmarks demonstrate that our method outperforms eight state-of-the-art image fusion methods in terms of both qualitative and quantitative evaluations. The code for our method, along with additional comparison results, will be made available at: https://github.com/QiaoYang-CV/SSPFUSION.
Authors: Xiang Gao, Hainan Cui, Yangdong Liu, Shuhan Shen
In order to further advance the accuracy and robustness of the incremental parameter estimation-based rotation averaging methods, in this paper, a new member of the Incremental Rotation Averaging (IRA) family is introduced, which is termed as IRAv4. As its most significant feature, a task-specific connected dominating set is extracted in IRAv4 to serve as a more reliable and accurate reference for rotation local-to-global alignment. This alignment reference is incrementally constructed, together with the absolute rotations of the vertices belong to it simultaneously estimated. Comprehensive evaluations are performed on the 1DSfM dataset, by which the effectiveness of both the reference construction method and the entire rotation averaging pipeline proposed in this paper is demonstrated.
Authors: Shilin Xu, Xiangtai Li, Size Wu, Wenwei Zhang, Yining Li, Guangliang Cheng, Yunhai Tong, Kai Chen, Chen Change Loy
Open-vocabulary object detection (OVOD) aims to detect the objects beyond the set of classes observed during training. This work presents a simple yet effective strategy that leverages the zero-shot classification ability of pre-trained vision-language models (VLM), such as CLIP, to directly discover proposals of possible novel classes. Unlike previous works that ignore novel classes during training and rely solely on the region proposal network (RPN) for novel object detection, our method selectively filters proposals based on specific design criteria. The resulting sets of identified proposals serve as pseudo-labels of potential novel classes during the training phase. This self-training strategy improves the recall and accuracy of novel classes without requiring additional annotations or datasets. We further propose a simple offline pseudo-label generation strategy to refine the object detector. Empirical evaluations on three datasets, including LVIS, V3Det, and COCO, demonstrate significant improvements over the baseline performance without incurring additional parameters or computational costs during inference. In particular, compared with previous F-VLM, our method achieves a 1.7\% improvement on the LVIS dataset. We also achieve over 6.5\% improvement on the recent challenging V3Det dataset. When combined with the recent method CLIPSelf, our method also achieves 46.7 novel class AP on COCO without introducing extra data for pertaining.
Authors: Christopher Hahne, Georges Chabouh, Arthur Chavignon, Olivier Couture, Raphael Sznitman
In Ultrasound Localization Microscopy (ULM),achieving high-resolution images relies on the precise localization of contrast agent particles across consecutive beam-formed frames. However, our study uncovers an enormous potential: The process of delay-and-sum beamforming leads to an irreversible reduction of Radio-Frequency (RF) data, while its implications for localization remain largely unexplored. The rich contextual information embedded within RF wavefronts, including their hyperbolic shape and phase, offers great promise for guiding Deep Neural Networks (DNNs) in challenging localization scenarios. To fully exploit this data, we propose to directly localize scatterers in RF signals. Our approach involves a custom super-resolution DNN using learned feature channel shuffling and a novel semi-global convolutional sampling block tailored for reliable and accurate wavefront localization. Additionally, we introduce a geometric point transformation that facilitates seamless mapping between RF and B-mode coordinate space. To understand the impact of beamforming on ULM, we validate the effectiveness of our method by conducting an extensive comparison with State-Of-The-Art (SOTA) techniques. We present the inaugural in vivo results from an RF-trained DNN, highlighting its real-world practicality. Our findings show that RF-ULM bridges the domain gap between synthetic and real datasets, offering a considerable advantage in terms of precision and complexity. To enable the broader research community to benefit from our findings, our code and the associated SOTA methods are made available at https://github.com/hahnec/rf-ulm.
Authors: Siyuan Cheng, Bozhong Tian, Qingbin Liu, Xi Chen, Yongheng Wang, Huajun Chen, Ningyu Zhang
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: Yangchao Wu, Tian Yu Liu, Hyoungseob Park, Stefano Soatto, Dong Lao, Alex Wong
Unsupervised depth completion methods are trained by minimizing sparse depth and image reconstruction error. Block artifacts from resampling, intensity saturation, and occlusions are amongst the many undesirable by-products of common data augmentation schemes that affect image reconstruction quality, and thus the training signal. Hence, typical augmentations on images viewed as essential to training pipelines in other vision tasks have seen limited use beyond small image intensity changes and flipping. The sparse depth modality have seen even less as intensity transformations alter the scale of the 3D scene, and geometric transformations may decimate the sparse points during resampling. We propose a method that unlocks a wide range of previously-infeasible geometric augmentations for unsupervised depth completion. This is achieved by reversing, or ``undo"-ing, geometric transformations to the coordinates of the output depth, warping the depth map back to the original reference frame. This enables computing the reconstruction losses using the original images and sparse depth maps, eliminating the pitfalls of naive loss computation on the augmented inputs. This simple yet effective strategy allows us to scale up augmentations to boost performance. We demonstrate our method on indoor (VOID) and outdoor (KITTI) datasets where we improve upon three existing methods by an average of 11.75% across both datasets.
Authors: Hao Lu, Yunpeng Zhang, Qing Lian, Dalong Du, Yingcong Chen
Detecting objects in 3D space using multiple cameras, known as Multi-Camera 3D Object Detection (MC3D-Det), has gained prominence with the advent of bird's-eye view (BEV) approaches. However, these methods often struggle when faced with unfamiliar testing environments due to the lack of diverse training data encompassing various viewpoints and environments. To address this, we propose a novel method that aligns 3D detection with 2D camera plane results, ensuring consistent and accurate detections. Our framework, anchored in perspective debiasing, helps the learning of features resilient to domain shifts. In our approach, we render diverse view maps from BEV features and rectify the perspective bias of these maps, leveraging implicit foreground volumes to bridge the camera and BEV planes. This two-step process promotes the learning of perspective- and context-independent features, crucial for accurate object detection across varying viewpoints, camera parameters, and environmental conditions. Notably, our model-agnostic approach preserves the original network structure without incurring additional inference costs, facilitating seamless integration across various models and simplifying deployment. Furthermore, we also show our approach achieves satisfactory results in real data when trained only with virtual datasets, eliminating the need for real scene annotations. Experimental results on both Domain Generalization (DG) and Unsupervised Domain Adaptation (UDA) clearly demonstrate its effectiveness. The codes are available at https://github.com/EnVision-Research/Generalizable-BEV.
Authors: Xiaohui Liu, Zhilu Zhang, Xiaohe Wu, Chaoyu Feng, Xiaotao Wang, Lei Lei, Wangmeng Zuo
Real-world image de-weathering aims at removing various undesirable weather-related artifacts. Owing to the impossibility of capturing image pairs concurrently, existing real-world de-weathering datasets often exhibit inconsistent illumination, position, and textures between the ground-truth images and the input degraded images, resulting in imperfect supervision. Such non-ideal supervision negatively affects the training process of learning-based de-weathering methods. In this work, we attempt to address the problem with a unified solution for various inconsistencies. Specifically, inspired by information bottleneck theory, we first develop a Consistent Label Constructor (CLC) to generate a pseudo-label as consistent as possible with the input degraded image while removing most weather-related degradations. In particular, multiple adjacent frames of the current input are also fed into CLC to enhance the pseudo-label. Then we combine the original imperfect labels and pseudo-labels to jointly supervise the de-weathering model by the proposed Information Allocation Strategy (IAS). During testing, only the de-weathering model is used for inference. Experiments on two real-world de-weathering datasets show that our method helps existing de-weathering models achieve better performance. Codes are available at https://github.com/1180300419/imperfect-deweathering.
Authors: Xingchen Zhao, Niluthpol Chowdhury Mithun, Abhinav Rajvanshi, Han-Pang Chiu, Supun Samarasekera
Deep learning-based solutions for semantic segmentation suffer from significant performance degradation when tested on data with different characteristics than what was used during the training. Adapting the models using annotated data from the new domain is not always practical. Unsupervised Domain Adaptation (UDA) approaches are crucial in deploying these models in the actual operating conditions. Recent state-of-the-art (SOTA) UDA methods employ a teacher-student self-training approach, where a teacher model is used to generate pseudo-labels for the new data which in turn guide the training process of the student model. Though this approach has seen a lot of success, it suffers from the issue of noisy pseudo-labels being propagated in the training process. To address this issue, we propose an auxiliary pseudo-label refinement network (PRN) for online refining of the pseudo labels and also localizing the pixels whose predicted labels are likely to be noisy. Being able to improve the quality of pseudo labels and select highly reliable ones, PRN helps self-training of segmentation models to be robust against pseudo label noise propagation during different stages of adaptation. We evaluate our approach on benchmark datasets with three different domain shifts, and our approach consistently performs significantly better than the previous state-of-the-art methods.
Authors: Seongsu Bae, Daeun Kyung, Jaehee Ryu, Eunbyeol Cho, Gyubok Lee, Sunjun Kweon, Jungwoo Oh, Lei Ji, Eric I-Chao Chang, Tackeun Kim, Edward Choi
Electronic Health Records (EHRs), which contain patients' medical histories in various multi-modal formats, often overlook the potential for joint reasoning across imaging and table modalities underexplored in current EHR Question Answering (QA) systems. In this paper, we introduce EHRXQA, a novel multi-modal question answering dataset combining structured EHRs and chest X-ray images. To develop our dataset, we first construct two uni-modal resources: 1) The MIMIC-CXR-VQA dataset, our newly created medical visual question answering (VQA) benchmark, specifically designed to augment the imaging modality in EHR QA, and 2) EHRSQL (MIMIC-IV), a refashioned version of a previously established table-based EHR QA dataset. By integrating these two uni-modal resources, we successfully construct a multi-modal EHR QA dataset that necessitates both uni-modal and cross-modal reasoning. To address the unique challenges of multi-modal questions within EHRs, we propose a NeuralSQL-based strategy equipped with an external VQA API. This pioneering endeavor enhances engagement with multi-modal EHR sources and we believe that our dataset can catalyze advances in real-world medical scenarios such as clinical decision-making and research. EHRXQA is available at https://github.com/baeseongsu/ehrxqa.
Authors: Chun Liu, Longwei Yang, Zheng Li, Wei Yang, Zhigang Han, Jianzhong Guo, Junyong Yu
Cross-domain few-shot hyperspectral image classification focuses on learning prior knowledge from a large number of labeled samples from source domains and then transferring the knowledge to the tasks which contain few labeled samples in target domains. Following the metric-based manner, many current methods first extract the features of the query and support samples, and then directly predict the classes of query samples according to their distance to the support samples or prototypes. The relations between samples have not been fully explored and utilized. Different from current works, this paper proposes to learn sample relations on different levels and take them into the model learning process, to improve the cross-domain few-shot hyperspectral image classification. Building on current method of "Deep Cross-Domain Few-Shot Learning for Hyperspectral Image Classification" which adopts a domain discriminator to deal with domain-level distribution difference, the proposed method applies contrastive learning to learn the class-level sample relations to obtain more discriminable sample features. In addition, it adopts a transformer based cross-attention learning module to learn the set-level sample relations and acquire the attention from query samples to support samples. Our experimental results have demonstrated the contribution of the multi-level relation learning mechanism for few-shot hyperspectral image classification when compared with the state of the art methods.
Authors: Yuanxin Liu, Lei Li, Shuhuai Ren, Rundong Gao, Shicheng Li, Sishuo Chen, Xu Sun, Lu Hou
Recently, open-domain text-to-video (T2V) generation models have made remarkable progress. However, the promising results are mainly shown by the qualitative cases of generated videos, while the quantitative evaluation of T2V models still faces two critical problems. Firstly, existing studies lack fine-grained evaluation of T2V models on different categories of text prompts. Although some benchmarks have categorized the prompts, their categorization either only focuses on a single aspect or fails to consider the temporal information in video generation. Secondly, it is unclear whether the automatic evaluation metrics are consistent with human standards. To address these problems, we propose FETV, a benchmark for Fine-grained Evaluation of Text-to-Video generation. FETV is multi-aspect, categorizing the prompts based on three orthogonal aspects: the major content, the attributes to control and the prompt complexity. FETV is also temporal-aware, which introduces several temporal categories tailored for video generation. Based on FETV, we conduct comprehensive manual evaluations of four representative T2V models, revealing their pros and cons on different categories of prompts from different aspects. We also extend FETV as a testbed to evaluate the reliability of automatic T2V metrics. The multi-aspect categorization of FETV enables fine-grained analysis of the metrics' reliability in different scenarios. We find that existing automatic metrics (e.g., CLIPScore and FVD) correlate poorly with human evaluation. To address this problem, we explore several solutions to improve CLIPScore and FVD, and develop two automatic metrics that exhibit significant higher correlation with humans than existing metrics. Benchmark page: https://github.com/llyx97/FETV.
Authors: Dasol Choi, Dongbin Na
Machine unlearning is a crucial tool for enabling a classification model to forget specific data that are used in the training time. Recently, various studies have presented machine unlearning algorithms and evaluated their methods on several datasets. However, most of the current machine unlearning algorithms have been evaluated solely on traditional computer vision datasets such as CIFAR-10, MNIST, and SVHN. Furthermore, previous studies generally evaluate the unlearning methods in the class-unlearning setup. Most previous work first trains the classification models and then evaluates the machine unlearning performance of machine unlearning algorithms by forgetting selected image classes (categories) in the experiments. Unfortunately, these class-unlearning settings might not generalize to real-world scenarios. In this work, we propose a machine unlearning setting that aims to unlearn specific instance that contains personal privacy (identity) while maintaining the original task of a given model. Specifically, we propose two machine unlearning benchmark datasets, MUFAC and MUCAC, that are greatly useful to evaluate the performance and robustness of a machine unlearning algorithm. In our benchmark datasets, the original model performs facial feature recognition tasks: face age estimation (multi-class classification) and facial attribute classification (binary class classification), where a class does not depend on any single target subject (personal identity), which can be a realistic setting. Moreover, we also report the performance of the state-of-the-art machine unlearning methods on our proposed benchmark datasets. All the datasets, source codes, and trained models are publicly available at https://github.com/ndb796/MachineUnlearning.
Authors: Fabiha Bushra, Muhammad E. H. Chowdhury, Rusab Sarmun, Saidul Kabir, Menatalla Said, Sohaib Bassam Zoghoul, Adam Mushtak, Israa Al-Hashimi, Abdulrahman Alqahtani, Anwarul Hasan
The increasing reliance on Computed Tomography Pulmonary Angiography for Pulmonary Embolism (PE) diagnosis presents challenges and a pressing need for improved diagnostic solutions. The primary objective of this study is to leverage deep learning techniques to enhance the Computer Assisted Diagnosis of PE. With this aim, we propose a classifier-guided detection approach that effectively leverages the classifier's probabilistic inference to direct the detection predictions, marking a novel contribution in the domain of automated PE diagnosis. Our end-to-end classification framework introduces an Attention-Guided Convolutional Neural Network (AG-CNN) that leverages local context by utilizing an attention mechanism. This approach emulates a human expert's attention by looking at both global appearances and local lesion regions before forming a conclusive decision. The classifier demonstrates strong performance on the FUMPE dataset, achieving AUROC, sensitivity, specificity, and F1-score of 0.927, 0.862, 0.879, and 0.805 respectively with Inception-v3 backbone architecture. Moreover, AG-CNN outperforms the baseline DenseNet-121 model, achieving an 8.1% AUROC gain. While prior studies have primarily focused on PE detection in main arteries, our utilization of cutting-edge object detection models and ensembling techniques greatly improves the accuracy of finding small embolisms in the peripheral arteries. Finally, our proposed classifier-guided detection approach further refines the detection metrics contributing new state-of-the-art to the community: mAP$_{50}$, sensitivity and F1-score of 0.846, 0.901 and 0.779 respectively outperforming the former benchmark with a significant 3.7% improvement in mAP$_{50}$. Our research aims to elevate PE patient care by integrating AI solutions into clinical workflows, highlighting the potential of human-AI collaboration in medical diagnostics.
Authors: Haokun Zhu, Juang Ian Chong, Teng Hu, Ran Yi, Yu-Kun Lai, Paul L. Rosin
Vector graphics are widely used in graphical designs and have received more and more attention. However, unlike raster images which can be easily obtained, acquiring high-quality vector graphics, typically through automatically converting from raster images remains a significant challenge, especially for more complex images such as photos or artworks. In this paper, we propose SAMVG, a multi-stage model to vectorize raster images into SVG (Scalable Vector Graphics). Firstly, SAMVG uses general image segmentation provided by the Segment-Anything Model and uses a novel filtering method to identify the best dense segmentation map for the entire image. Secondly, SAMVG then identifies missing components and adds more detailed components to the SVG. Through a series of extensive experiments, we demonstrate that SAMVG can produce high quality SVGs in any domain while requiring less computation time and complexity compared to previous state-of-the-art methods.
Authors: Yunlong Zhang, Honglin Li, Yuxuan Sun, Sunyi Zheng, Chenglu Zhu, Lin Yang
Overfitting is a significant challenge in the application of Multiple Instance Learning (MIL) methods for Whole Slide Image (WSI) analysis. Visualizing attention heatmaps reveals that current MIL methods focus on a subset of discriminative instances, hindering effective model generalization. To tackle this, we propose Attention-Challenging MIL (ACMIL), aimed at forcing the attention mechanism to focus on more challenging instances. ACMIL incorporates two techniques, Multiple Branch Attention (MBA) to capture more discriminative instances and Stochastic Top-K Instance Masking (STKIM) to suppress top-k salient instances. Evaluation on three WSI datasets with two pre-trained backbones outperforms state-of-the-art methods. Additionally, through heatmap visualization and UMAP visualization, this paper comprehensively illustrates ACMIL's effectiveness in overcoming the overfitting challenge. The source code is available at \url{https://github.com/dazhangyu123/ACMIL}.
Authors: Rui Cai, Xuying Ning, Peter N. Belhumeur
In the post-pandemic era, wearing face masks has posed great challenge to the ordinary face recognition. In the previous study, researchers has applied pretrained VGG16, and ResNet50 to extract features on the elaborate curated existing masked face recognition (MFR) datasets, RMFRD and SMFRD. To make the model more adaptable to the real world situation where the sample size is smaller and the camera environment has greater changes, we created a more challenging masked face dataset ourselves, by selecting 50 identities with 1702 images from Labelled Faces in the Wild (LFW) Dataset, and simulated face masks through key point detection. The another part of our study is to solve the masked face recognition problem, and we chose models by referring to the former state of the art results, instead of directly using pretrained models, we fine tuned the model on our new dataset and use the last linear layer to do the classification directly. Furthermore, we proposed using data augmentation strategy to further increase the test accuracy, and fine tuned a new networks beyond the former study, one of the most SOTA networks, Inception ResNet v1. The best test accuracy on 50 identity MFR has achieved 95%.
Authors: Jiaming Zhang, Xingjun Ma, Xin Wang, Lingyu Qiu, Jiaqi Wang, Yu-Gang Jiang, Jitao Sang
With the rapid advancement of multimodal learning, pre-trained Vision-Language Models (VLMs) such as CLIP have demonstrated remarkable capacities in bridging the gap between visual and language modalities. However, these models remain vulnerable to adversarial attacks, particularly in the image modality, presenting considerable security risks. This paper introduces Adversarial Prompt Tuning (AdvPT), a novel technique to enhance the adversarial robustness of image encoders in VLMs. AdvPT innovatively leverages learnable text prompts and aligns them with adversarial image embeddings, to address the vulnerabilities inherent in VLMs without the need for extensive parameter training or modification of the model architecture. We demonstrate that AdvPT improves resistance against white-box and black-box adversarial attacks and exhibits a synergistic effect when combined with existing image-processing-based defense techniques, further boosting defensive capabilities. Comprehensive experimental analyses provide insights into adversarial prompt tuning, a novel paradigm devoted to improving resistance to adversarial images through textual input modifications, paving the way for future robust multimodal learning research. These findings open up new possibilities for enhancing the security of VLMs. Our code is available at https://github.com/jiamingzhang94/Adversarial-Prompt-Tuning.
Authors: Shutong Zhang, Yi-Ling Qiao, Guanglei Zhu, Eric Heiden, Dylan Turpin, Jingzhou Liu, Ming Lin, Miles Macklin, Animesh Garg
Various heuristic objectives for modeling hand-object interaction have been proposed in past work. However, due to the lack of a cohesive framework, these objectives often possess a narrow scope of applicability and are limited by their efficiency or accuracy. In this paper, we propose HandyPriors, a unified and general pipeline for pose estimation in human-object interaction scenes by leveraging recent advances in differentiable physics and rendering. Our approach employs rendering priors to align with input images and segmentation masks along with physics priors to mitigate penetration and relative-sliding across frames. Furthermore, we present two alternatives for hand and object pose estimation. The optimization-based pose estimation achieves higher accuracy, while the filtering-based tracking, which utilizes the differentiable priors as dynamics and observation models, executes faster. We demonstrate that HandyPriors attains comparable or superior results in the pose estimation task, and that the differentiable physics module can predict contact information for pose refinement. We also show that our approach generalizes to perception tasks, including robotic hand manipulation and human-object pose estimation in the wild.
Authors: Lingteng Qiu, Guanying Chen, Xiaodong Gu, Qi Zuo, Mutian Xu, Yushuang Wu, Weihao Yuan, Zilong Dong, Liefeng Bo, Xiaoguang Han
Lifting 2D diffusion for 3D generation is a challenging problem due to the lack of geometric prior and the complex entanglement of materials and lighting in natural images. Existing methods have shown promise by first creating the geometry through score-distillation sampling (SDS) applied to rendered surface normals, followed by appearance modeling. However, relying on a 2D RGB diffusion model to optimize surface normals is suboptimal due to the distribution discrepancy between natural images and normals maps, leading to instability in optimization. In this paper, recognizing that the normal and depth information effectively describe scene geometry and be automatically estimated from images, we propose to learn a generalizable Normal-Depth diffusion model for 3D generation. We achieve this by training on the large-scale LAION dataset together with the generalizable image-to-depth and normal prior models. In an attempt to alleviate the mixed illumination effects in the generated materials, we introduce an albedo diffusion model to impose data-driven constraints on the albedo component. Our experiments show that when integrated into existing text-to-3D pipelines, our models significantly enhance the detail richness, achieving state-of-the-art results. Our project page is https://aigc3d.github.io/richdreamer/.
Authors: Wang Zhu, Ishika Singh, Yuan Huang, Robin Jia, Jesse Thomason
Data augmentation via back-translation is common when pretraining Vision-and-Language Navigation (VLN) models, even though the generated instructions are noisy. But: does that noise matter? We find that nonsensical or irrelevant language instructions during pretraining can have little effect on downstream performance for both HAMT and VLN-BERT on R2R, and is still better than only using clean, human data. To underscore these results, we concoct an efficient augmentation method, Unigram + Object, which generates nonsensical instructions that nonetheless improve downstream performance. Our findings suggest that what matters for VLN R2R pretraining is the quantity of visual trajectories, not the quality of instructions.
Authors: Yilin Wen, Hao Pan, Takehiko Ohkawa, Lei Yang, Jia Pan, Yoichi Sato, Taku Komura, Wenping Wang
We present a novel framework that concurrently tackles hand action recognition and 3D future hand motion prediction. While previous works focus on either recognition or prediction, we propose a generative Transformer VAE architecture to jointly capture both aspects, facilitating realistic motion prediction by leveraging the short-term hand motion and long-term action consistency observed across timestamps. To ensure faithful representation of the semantic dependency and different temporal granularity of hand pose and action, our framework is decomposed into two cascaded VAE blocks. The lower pose block models short-span poses, while the upper action block models long-span action. These are connected by a mid-level feature that represents sub-second series of hand poses. Our framework is trained across multiple datasets, where pose and action blocks are trained separately to fully utilize pose-action annotations of different qualities. Evaluations show that on multiple datasets, the joint modeling of recognition and prediction improves over separate solutions, and the semantic and temporal hierarchy enables long-term pose and action modeling.
Authors: Yingdong Hu, Fanqi Lin, Tong Zhang, Li Yi, Yang Gao
In this study, we are interested in imbuing robots with the capability of physically-grounded task planning. Recent advancements have shown that large language models (LLMs) possess extensive knowledge useful in robotic tasks, especially in reasoning and planning. However, LLMs are constrained by their lack of world grounding and dependence on external affordance models to perceive environmental information, which cannot jointly reason with LLMs. We argue that a task planner should be an inherently grounded, unified multimodal system. To this end, we introduce Robotic Vision-Language Planning (ViLa), a novel approach for long-horizon robotic planning that leverages vision-language models (VLMs) to generate a sequence of actionable steps. ViLa directly integrates perceptual data into its reasoning and planning process, enabling a profound understanding of commonsense knowledge in the visual world, including spatial layouts and object attributes. It also supports flexible multimodal goal specification and naturally incorporates visual feedback. Our extensive evaluation, conducted in both real-robot and simulated environments, demonstrates ViLa's superiority over existing LLM-based planners, highlighting its effectiveness in a wide array of open-world manipulation tasks.
Authors: Xianda Guo, Juntao Lu, Chenming Zhang, Yiqi Wang, Yiqun Duan, Tian Yang, Zheng Zhu, Long Chen
Stereo matching, a pivotal technique in computer vision, plays a crucial role in robotics, autonomous navigation, and augmented reality. Despite the development of numerous impressive methods in recent years, replicating their results and determining the most suitable architecture for practical application remains challenging. Addressing this gap, our paper introduces a comprehensive benchmark focusing on practical applicability rather than solely on performance enhancement. Specifically, we develop a flexible and efficient stereo matching codebase, called OpenStereo. OpenStereo includes training and inference codes of more than 12 network models, making it, to our knowledge, the most complete stereo matching toolbox available. Based on OpenStereo, we conducted experiments on the SceneFlow dataset and have achieved or surpassed the performance metrics reported in the original paper. Additionally, we conduct an in-depth revisitation of recent developments in stereo matching through ablative experiments. These investigations inspired the creation of StereoBase, a simple yet strong baseline model. Our extensive comparative analyses of StereoBase against numerous contemporary stereo matching methods on the SceneFlow dataset demonstrate its remarkably strong performance. The source code is available at https://github.com/XiandaGuo/OpenStereo.
Authors: Rui Huang (1), Binbin Jiang (1), Qingyi Zhao (1), William Wang, Yuxiang Zhang (1), Qing Guo (2 and 3) ((1) College of Computer Science and Technology, Civil Aviation University of China, China, (2) IHPC, Agency for Science, Technology and Research, Singapore, (3) CFAR, Agency for Science, Technology and Research, Singapore)
In this work, we aim to detect the changes caused by object variations in a scene represented by the neural radiance fields (NeRFs). Given an arbitrary view and two sets of scene images captured at different timestamps, we can predict the scene changes in that view, which has significant potential applications in scene monitoring and measuring. We conducted preliminary studies and found that such an exciting task cannot be easily achieved by utilizing existing NeRFs and 2D change detection methods with many false or missing detections. The main reason is that the 2D change detection is based on the pixel appearance difference between spatial-aligned image pairs and neglects the stereo information in the NeRF. To address the limitations, we propose the C-NERF to represent scene changes as directional consistency difference-based NeRF, which mainly contains three modules. We first perform the spatial alignment of two NeRFs captured before and after changes. Then, we identify the change points based on the direction-consistent constraint; that is, real change points have similar change representations across view directions, but fake change points do not. Finally, we design the change map rendering process based on the built NeRFs and can generate the change map of an arbitrarily specified view direction. To validate the effectiveness, we build a new dataset containing ten scenes covering diverse scenarios with different changing objects. Our approach surpasses state-of-the-art 2D change detection and NeRF-based methods by a significant margin.
Authors: Yuexing Han, Guanxin Wan, Bing Wang
Deep learning has yielded remarkable outcomes in various domains. However, the challenge of requiring large-scale labeled samples still persists in deep learning. Thus, data augmentation has been introduced as a critical strategy to train deep learning models. However, data augmentation suffers from information loss and poor performance in small sample environments. To overcome these drawbacks, we propose a feature augmentation method based on shape space theory, i.e., feature augmentation on Geodesic curve, called FAGC in brevity.First, we extract features from the image with the neural network model. Then, the multiple image features are projected into a pre-shape space as features. In the pre-shape space, a Geodesic curve is built to fit the features. Finally, the many generated features on the Geodesic curve are used to train the various machine learning models. The FAGC module can be seamlessly integrated with most machine learning methods. And the proposed method is simple, effective and insensitive for the small sample datasets.Several examples demonstrate that the FAGC method can greatly improve the performance of the data preprocessing model in a small sample environment.
Authors: Shangbo Wu, Yu-an Tan, Yajie Wang, Ruinan Ma, Wencong Ma, Yuanzhang Li
Adversarial transferability enables black-box attacks on unknown victim deep neural networks (DNNs), rendering attacks viable in real-world scenarios. Current transferable attacks create adversarial perturbation over the entire image, resulting in excessive noise that overfit the source model. Concentrating perturbation to dominant image regions that are model-agnostic is crucial to improving adversarial efficacy. However, limiting perturbation to local regions in the spatial domain proves inadequate in augmenting transferability. To this end, we propose a transferable adversarial attack with fine-grained perturbation optimization in the frequency domain, creating centralized perturbation. We devise a systematic pipeline to dynamically constrain perturbation optimization to dominant frequency coefficients. The constraint is optimized in parallel at each iteration, ensuring the directional alignment of perturbation optimization with model prediction. Our approach allows us to centralize perturbation towards sample-specific important frequency features, which are shared by DNNs, effectively mitigating source model overfitting. Experiments demonstrate that by dynamically centralizing perturbation on dominating frequency coefficients, crafted adversarial examples exhibit stronger transferability, and allowing them to bypass various defenses.
Authors: Hansong Zhang, Shikun Li, Dan Zeng, Chenggang Yan, Shiming Ge
As the size of the datasets getting larger, accurately annotating such datasets is becoming more impractical due to the expensiveness on both time and economy. Therefore, crowd-sourcing has been widely adopted to alleviate the cost of collecting labels, which also inevitably introduces label noise and eventually degrades the performance of the model. To learn from crowd-sourcing annotations, modeling the expertise of each annotator is a common but challenging paradigm, because the annotations collected by crowd-sourcing are usually highly-sparse. To alleviate this problem, we propose Coupled Confusion Correction (CCC), where two models are simultaneously trained to correct the confusion matrices learned by each other. Via bi-level optimization, the confusion matrices learned by one model can be corrected by the distilled data from the other. Moreover, we cluster the ``annotator groups'' who share similar expertise so that their confusion matrices could be corrected together. In this way, the expertise of the annotators, especially of those who provide seldom labels, could be better captured. Remarkably, we point out that the annotation sparsity not only means the average number of labels is low, but also there are always some annotators who provide very few labels, which is neglected by previous works when constructing synthetic crowd-sourcing annotations. Based on that, we propose to use Beta distribution to control the generation of the crowd-sourcing labels so that the synthetic annotations could be more consistent with the real-world ones. Extensive experiments are conducted on two types of synthetic datasets and three real-world datasets, the results of which demonstrate that CCC significantly outperforms state-of-the-art approaches.
Authors: Zihao Zhao, Yuxiao Liu, Han Wu, Yonghao Li, Sheng Wang, Lin Teng, Disheng Liu, Zhiming Cui, Qian Wang, Dinggang Shen
Contrastive Language-Image Pre-training (CLIP), a simple yet effective pre-training paradigm, successfully introduces text supervision to vision models. It has shown promising results across various tasks, attributable to its generalizability and interpretability. The use of CLIP has recently gained increasing interest in the medical imaging domain, serving both as a pre-training paradigm for aligning medical vision and language, and as a critical component in diverse clinical tasks. With the aim of facilitating a deeper understanding of this promising direction, this survey offers an in-depth exploration of the CLIP paradigm within the domain of medical imaging, regarding both refined CLIP pre-training and CLIP-driven applications. In this study, We (1) start with a brief introduction to the fundamentals of CLIP methodology. (2) Then, we investigate the adaptation of CLIP pre-training in the medical domain, focusing on how to optimize CLIP given characteristics of medical images and reports. (3) Furthermore, we explore the practical utilization of CLIP pre-trained models in various tasks, including classification, dense prediction, and cross-modal tasks. (4) Finally, we discuss existing limitations of CLIP in the context of medical imaging and propose forward-looking directions to address the demands of medical imaging domain. We expect that this comprehensive survey will provide researchers in the field of medical image analysis with a holistic understanding of the CLIP paradigm and its potential implications. The project page can be found on https://github.com/zhaozh10/Awesome-CLIP-in-Medical-Imaging.
Authors: Qi Tang, Yao Zhao, Meiqin Liu, Jian Jin, Chao Yao
As a critical clue of video super-resolution (VSR), inter-frame alignment significantly impacts overall performance. However, accurate pixel-level alignment is a challenging task due to the intricate motion interweaving in the video. In response to this issue, we introduce a novel paradigm for VSR named Semantic Lens, predicated on semantic priors drawn from degraded videos. Specifically, video is modeled as instances, events, and scenes via a Semantic Extractor. Those semantics assist the Pixel Enhancer in understanding the recovered contents and generating more realistic visual results. The distilled global semantics embody the scene information of each frame, while the instance-specific semantics assemble the spatial-temporal contexts related to each instance. Furthermore, we devise a Semantics-Powered Attention Cross-Embedding (SPACE) block to bridge the pixel-level features with semantic knowledge, composed of a Global Perspective Shifter (GPS) and an Instance-Specific Semantic Embedding Encoder (ISEE). Concretely, the GPS module generates pairs of affine transformation parameters for pixel-level feature modulation conditioned on global semantics. After that, the ISEE module harnesses the attention mechanism to align the adjacent frames in the instance-centric semantic space. In addition, we incorporate a simple yet effective pre-alignment module to alleviate the difficulty of model training. Extensive experiments demonstrate the superiority of our model over existing state-of-the-art VSR methods.
Authors: Zhiyuan Ma, Guoli Jia, Bowen Zhou
With the great success of text-conditioned diffusion models in creative text-to-image generation, various text-driven image editing approaches have attracted the attentions of many researchers. However, previous works mainly focus on discreteness-sensitive instructions such as adding, removing or replacing specific objects, background elements or global styles (i.e., hard editing), while generally ignoring subject-binding but semantically fine-changing continuity-sensitive instructions such as actions, poses or adjectives, and so on (i.e., soft editing), which hampers generative AI from generating user-customized visual contents. To mitigate this predicament, we propose a spatio-temporal guided adaptive editing algorithm AdapEdit, which realizes adaptive image editing by introducing a soft-attention strategy to dynamically vary the guiding degree from the editing conditions to visual pixels from both temporal and spatial perspectives. Note our approach has a significant advantage in preserving model priors and does not require model training, fine-tuning, extra data, or optimization. We present our results over a wide variety of raw images and editing instructions, demonstrating competitive performance and showing it significantly outperforms the previous approaches.
Authors: Haoyu Guo, He Zhu, Sida Peng, Yuang Wang, Yujun Shen, Ruizhen Hu, Xiaowei Zhou
This paper addresses the challenge of 3D instance segmentation by simultaneously leveraging 3D geometric and multi-view image information. Many previous works have applied deep learning techniques to 3D point clouds for instance segmentation. However, these methods often failed to generalize to various types of scenes due to the scarcity and low-diversity of labeled 3D point cloud data. Some recent works have attempted to lift 2D instance segmentations to 3D within a bottom-up framework. The inconsistency in 2D instance segmentations among views can substantially degrade the performance of 3D segmentation. In this work, we introduce a novel 3D-to-2D query framework to effectively exploit 2D segmentation models for 3D instance segmentation. Specifically, we pre-segment the scene into several superpoints in 3D, formulating the task into a graph cut problem. The superpoint graph is constructed based on 2D segmentation models, where node features are obtained from multi-view image features and edge weights are computed based on multi-view segmentation results, enabling the better generalization ability. To process the graph, we train a graph neural network using pseudo 3D labels from 2D segmentation models. Experimental results on the ScanNet, ScanNet++ and KITTI-360 datasets demonstrate that our method achieves robust segmentation performance and can generalize across different types of scenes. Our project page is available at https://zju3dv.github.io/sam_graph.
Authors: Dongchen Han, Tianzhu Ye, Yizeng Han, Zhuofan Xia, Shiji Song, Gao Huang
The attention module is the key component in Transformers. While the global attention mechanism offers high expressiveness, its excessive computational cost restricts its applicability in various scenarios. In this paper, we propose a novel attention paradigm, Agent Attention, to strike a favorable balance between computational efficiency and representation power. Specifically, the Agent Attention, denoted as a quadruple $(Q, A, K, V)$, introduces an additional set of agent tokens $A$ into the conventional attention module. The agent tokens first act as the agent for the query tokens $Q$ to aggregate information from $K$ and $V$, and then broadcast the information back to $Q$. Given the number of agent tokens can be designed to be much smaller than the number of query tokens, the agent attention is significantly more efficient than the widely adopted Softmax attention, while preserving global context modelling capability. Interestingly, we show that the proposed agent attention is equivalent to a generalized form of linear attention. Therefore, agent attention seamlessly integrates the powerful Softmax attention and the highly efficient linear attention. Extensive experiments demonstrate the effectiveness of agent attention with various vision Transformers and across diverse vision tasks, including image classification, object detection, semantic segmentation and image generation. Notably, agent attention has shown remarkable performance in high-resolution scenarios, owning to its linear attention nature. For instance, when applied to Stable Diffusion, our agent attention accelerates generation and substantially enhances image generation quality without any additional training. Code is available at https://github.com/LeapLabTHU/Agent-Attention.
Authors: Yuanyou Xu, Zongxin Yang, Yi Yang
Powered by large-scale text-to-image generation models, text-to-3D avatar generation has made promising progress. However, most methods fail to produce photorealistic results, limited by imprecise geometry and low-quality appearance. Towards more practical avatar generation, we present SEEAvatar, a method for generating photorealistic 3D avatars from text with SElf-Evolving constraints for decoupled geometry and appearance. For geometry, we propose to constrain the optimized avatar in a decent global shape with a template avatar. The template avatar is initialized with human prior and can be updated by the optimized avatar periodically as an evolving template, which enables more flexible shape generation. Besides, the geometry is also constrained by the static human prior in local parts like face and hands to maintain the delicate structures. For appearance generation, we use diffusion model enhanced by prompt engineering to guide a physically based rendering pipeline to generate realistic textures. The lightness constraint is applied on the albedo texture to suppress incorrect lighting effect. Experiments show that our method outperforms previous methods on both global and local geometry and appearance quality by a large margin. Since our method can produce high-quality meshes and textures, such assets can be directly applied in classic graphics pipeline for realistic rendering under any lighting condition. Project page at: https://yoxu515.github.io/SEEAvatar/.
Authors: Wenhai Wang, Jiangwei Xie, ChuanYang Hu, Haoming Zou, Jianan Fan, Wenwen Tong, Yang Wen, Silei Wu, Hanming Deng, Zhiqi Li, Hao Tian, Lewei Lu, Xizhou Zhu, Xiaogang Wang, Yu Qiao, Jifeng Dai
Large language models (LLMs) have opened up new possibilities for intelligent agents, endowing them with human-like thinking and cognitive abilities. In this work, we delve into the potential of large language models (LLMs) in autonomous driving (AD). We introduce DriveMLM, an LLM-based AD framework that can perform close-loop autonomous driving in realistic simulators. To this end, (1) we bridge the gap between the language decisions and the vehicle control commands by standardizing the decision states according to the off-the-shelf motion planning module. (2) We employ a multi-modal LLM (MLLM) to model the behavior planning module of a module AD system, which uses driving rules, user commands, and inputs from various sensors (e.g., camera, lidar) as input and makes driving decisions and provide explanations; This model can plug-and-play in existing AD systems such as Apollo for close-loop driving. (3) We design an effective data engine to collect a dataset that includes decision state and corresponding explanation annotation for model training and evaluation. We conduct extensive experiments and show that our model achieves 76.1 driving score on the CARLA Town05 Long, and surpasses the Apollo baseline by 4.7 points under the same settings, demonstrating the effectiveness of our model. We hope this work can serve as a baseline for autonomous driving with LLMs. Code and models shall be released at https://github.com/OpenGVLab/DriveMLM.
Authors: Chen Ma, Ningfei Wang, Qi Alfred Chen, Chao Shen
In Autonomous Driving (AD), real-time perception is a critical component responsible for detecting surrounding objects to ensure safe driving. While researchers have extensively explored the integrity of AD perception due to its safety and security implications, the aspect of availability (real-time performance) or latency has received limited attention. Existing works on latency-based attack have focused mainly on object detection, i.e., a component in camera-based AD perception, overlooking the entire camera-based AD perception, which hinders them to achieve effective system-level effects, such as vehicle crashes. In this paper, we propose SlowTrack, a novel framework for generating adversarial attacks to increase the execution time of camera-based AD perception. We propose a novel two-stage attack strategy along with the three new loss function designs. Our evaluation is conducted on four popular camera-based AD perception pipelines, and the results demonstrate that SlowTrack significantly outperforms existing latency-based attacks while maintaining comparable imperceptibility levels. Furthermore, we perform the evaluation on Baidu Apollo, an industry-grade full-stack AD system, and LGSVL, a production-grade AD simulator, with two scenarios to compare the system-level effects of SlowTrack and existing attacks. Our evaluation results show that the system-level effects can be significantly improved, i.e., the vehicle crash rate of SlowTrack is around 95% on average while existing works only have around 30%.
Authors: Bowen Tao, Lan Li, Xin-Chun Li, De-Chuan Zhan
Due to the advantages of leveraging unlabeled data and learning meaningful representations, semi-supervised learning and contrastive learning have been progressively combined to achieve better performances in popular applications with few labeled data and abundant unlabeled data. One common manner is assigning pseudo-labels to unlabeled samples and selecting positive and negative samples from pseudo-labeled samples to apply contrastive learning. However, the real-world data may be imbalanced, causing pseudo-labels to be biased toward the majority classes and further undermining the effectiveness of contrastive learning. To address the challenge, we propose Contrastive Learning with Augmented Features (CLAF). We design a class-dependent feature augmentation module to alleviate the scarcity of minority class samples in contrastive learning. For each pseudo-labeled sample, we select positive and negative samples from labeled data instead of unlabeled data to compute contrastive loss. Comprehensive experiments on imbalanced image classification datasets demonstrate the effectiveness of CLAF in the context of imbalanced semi-supervised learning.
Authors: Chandresh Pravin, Ivan Martino, Giuseppe Nicosia, Varun Ojha
We propose a systematic analysis of deep neural networks (DNNs) based on a signal processing technique for network parameter removal, in the form of synaptic filters that identifies the fragility, robustness and antifragility characteristics of DNN parameters. Our proposed analysis investigates if the DNN performance is impacted negatively, invariantly, or positively on both clean and adversarially perturbed test datasets when the DNN undergoes synaptic filtering. We define three \textit{filtering scores} for quantifying the fragility, robustness and antifragility characteristics of DNN parameters based on the performances for (i) clean dataset, (ii) adversarial dataset, and (iii) the difference in performances of clean and adversarial datasets. We validate the proposed systematic analysis on ResNet-18, ResNet-50, SqueezeNet-v1.1 and ShuffleNet V2 x1.0 network architectures for MNIST, CIFAR10 and Tiny ImageNet datasets. The filtering scores, for a given network architecture, identify network parameters that are invariant in characteristics across different datasets over learning epochs. Vice-versa, for a given dataset, the filtering scores identify the parameters that are invariant in characteristics across different network architectures. We show that our synaptic filtering method improves the test accuracy of ResNet and ShuffleNet models on adversarial datasets when only the robust and antifragile parameters are selectively retrained at any given epoch, thus demonstrating applications of the proposed strategy in improving model robustness.
Authors: Yuqian Yuan, Wentong Li, Jian Liu, Dongqi Tang, Xinjie Luo, Chi Qin, Lei Zhang, Jianke Zhu
Multimodal large language models (MLLMs) have recently achieved impressive general-purpose vision-language capabilities through visual instruction tuning. However, current MLLMs primarily focus on image-level or box-level understanding, falling short of achieving fine-grained vision-language alignment at the pixel level. Besides, the lack of mask-based instruction data limits their advancements. In this paper, we propose Osprey, a mask-text instruction tuning approach, to extend MLLMs by incorporating fine-grained mask regions into language instruction, aiming at achieving pixel-wise visual understanding. To achieve this goal, we first meticulously curate a mask-based region-text dataset with 724K samples, and then design a vision-language model by injecting pixel-level representation into LLM. Especially, Osprey adopts a convolutional CLIP backbone as the vision encoder and employs a mask-aware visual extractor to extract precise visual mask features from high resolution input. Experimental results demonstrate Osprey's superiority in various region understanding tasks, showcasing its new capability for pixel-level instruction tuning. In particular, Osprey can be integrated with Segment Anything Model (SAM) seamlessly to obtain multi-granularity semantics. The source code, dataset and demo can be found at https://github.com/CircleRadon/Osprey.
Authors: Yingzhe Peng, Xu Yang, Haoxuan Ma, Shuo Xu, Chi Zhang, Yucheng Han, Hanwang Zhang
This paper studies how to configure powerful In-Context Demonstration (ICD) sequences for a Large Vision-Language Model (LVLM) to solve Vision-Language tasks through In-Context Learning (ICL). After observing that configuring an ICD sequence is a mirror process of composing a sentence, i.e., just as a sentence can be composed word by word via a Language Model, an ICD sequence can also be configured one by one. Consequently, we introduce an ICD Language Model (ICD-LM) specifically designed to generate effective ICD sequences. This involves creating a dataset of hand-crafted ICD sequences for various query samples and using it to train the ICD-LM. Our approach, diverging from traditional methods in NLP that select and order ICDs separately, enables to simultaneously learn how to select and order ICDs, enhancing the effect of the sequences. Moreover, during data construction, we use the LVLM intended for ICL implementation to validate the strength of each ICD sequence, resulting in a model-specific dataset and the ICD-LM trained by this dataset is also model-specific. We validate our methodology through experiments in Visual Question Answering and Image Captioning, confirming the viability of using a Language Model for ICD configuration. Our comprehensive ablation studies further explore the impact of various dataset construction and ICD-LM development settings on the outcomes. The code is given in https://github.com/ForJadeForest/ICD-LM.
Authors: Wentao Wang, Xuanyao Huang, Tianyang Wang, Swalpa Kumar Roy
This paper explores the image synthesis capabilities of GPT-4, a leading multi-modal large language model. We establish a benchmark for evaluating the fidelity of texture features in images generated by GPT-4, comprising manually painted pictures and their AI-generated counterparts. The contributions of this study are threefold: First, we provide an in-depth analysis of the fidelity of image synthesis features based on GPT-4, marking the first such study on this state-of-the-art model. Second, the quantitative and qualitative experiments fully reveals the limitations of the GPT-4 model in image synthesis. Third, we have compiled a unique benchmark of manual drawings and corresponding GPT-4-generated images, introducing a new task to advance fidelity research in AI-generated content (AIGC). The dataset is available at: \url{https://github.com/rickwang28574/DeepArt}.
Authors: Congyi Wang
In this study, we introduce T2M-HiFiGPT, a novel conditional generative framework for synthesizing human motion from textual descriptions. This framework is underpinned by a Residual Vector Quantized Variational AutoEncoder (RVQ-VAE) and a double-tier Generative Pretrained Transformer (GPT) architecture. We demonstrate that our CNN-based RVQ-VAE is capable of producing highly accurate 2D temporal-residual discrete motion representations. Our proposed double-tier GPT structure comprises a temporal GPT and a residual GPT. The temporal GPT efficiently condenses information from previous frames and textual descriptions into a 1D context vector. This vector then serves as a context prompt for the residual GPT, which generates the final residual discrete indices. These indices are subsequently transformed back into motion data by the RVQ-VAE decoder. To mitigate the exposure bias issue, we employ straightforward code corruption techniques for RVQ and a conditional dropout strategy, resulting in enhanced synthesis performance. Remarkably, T2M-HiFiGPT not only simplifies the generative process but also surpasses existing methods in both performance and parameter efficacy, including the latest diffusion-based and GPT-based models. On the HumanML3D and KIT-ML datasets, our framework achieves exceptional results across nearly all primary metrics. We further validate the efficacy of our framework through comprehensive ablation studies on the HumanML3D dataset, examining the contribution of each component. Our findings reveal that RVQ-VAE is more adept at capturing precise 3D human motion with comparable computational demand compared to its VQ-VAE counterparts. As a result, T2M-HiFiGPT enables the generation of human motion with significantly increased accuracy, outperforming recent state-of-the-art approaches such as T2M-GPT and Att-T2M.
Authors: Jiankai Sun, Chuanyang Zheng, Enze Xie, Zhengying Liu, Ruihang Chu, Jianing Qiu, Jiaqi Xu, Mingyu Ding, Hongyang Li, Mengzhe Geng, Yue Wu, Wenhai Wang, Junsong Chen, Zhangyue Yin, Xiaozhe Ren, Jie Fu, Junxian He, Wu Yuan, Qi Liu, Xihui Liu, Yu Li, Hao Dong, Yu Cheng, Ming Zhang, Pheng Ann Heng, Jifeng Dai, Ping Luo, Jingdong Wang, Ji-Rong Wen, Xipeng Qiu, Yike Guo, Hui Xiong, Qun Liu, Zhenguo Li
Reasoning, a crucial ability for complex problem-solving, plays a pivotal role in various real-world settings such as negotiation, medical diagnosis, and criminal investigation. It serves as a fundamental methodology in the field of Artificial General Intelligence (AGI). With the ongoing development of foundation models, there is a growing interest in exploring their abilities in reasoning tasks. In this paper, we introduce seminal foundation models proposed or adaptable for reasoning, highlighting the latest advancements in various reasoning tasks, methods, and benchmarks. We then delve into the potential future directions behind the emergence of reasoning abilities within foundation models. We also discuss the relevance of multimodal learning, autonomous agents, and super alignment in the context of reasoning. By discussing these future research directions, we hope to inspire researchers in their exploration of this field, stimulate further advancements in reasoning with foundation models, and contribute to the development of AGI.
Authors: Alonso Urbano, David W. Romero
Group equivariance ensures consistent responses to group transformations of the input, leading to more robust models and enhanced generalization capabilities. However, this property can lead to overly constrained models if the symmetries considered in the group differ from those observed in data. While common methods address this by determining the appropriate level of symmetry at the dataset level, they are limited to supervised settings and ignore scenarios in which multiple levels of symmetry co-exist in the same dataset. For instance, pictures of cars and planes exhibit different levels of rotation, yet both are included in the CIFAR-10 dataset. In this paper, we propose a method able to detect the level of symmetry of each input without the need for labels. To this end, we derive a sufficient and necessary condition to learn the distribution of symmetries in the data. Using the learned distribution, we generate pseudo-labels that allow us to learn the levels of symmetry of each input in a self-supervised manner. We validate the effectiveness of our approach on synthetic datasets with different per-class levels of symmetries e.g. MNISTMultiple, in which digits are uniformly rotated within a class-dependent interval. We demonstrate that our method can be used for practical applications such as the generation of standardized datasets in which the symmetries are not present, as well as the detection of out-of-distribution symmetries during inference. By doing so, both the generalization and robustness of non-equivariant models can be improved. Our code is publicly available at https://github.com/aurban0/ssl-sym.
Authors: Ruiyuan Zhang, Jiaxiang Liu, Zexi Li, Hao Dong, Jie Fu, Chao Wu
Geometric fracture assembly presents a challenging practical task in archaeology and 3D computer vision. Previous methods have focused solely on assembling fragments based on semantic information, which has limited the quantity of objects that can be effectively assembled. Therefore, there is a need to develop a scalable framework for geometric fracture assembly without relying on semantic information. To improve the effectiveness of assembling geometric fractures without semantic information, we propose a co-creation space comprising several assemblers capable of gradually and unambiguously assembling fractures. Additionally, we introduce a novel loss function, i.e., the geometric-based collision loss, to address collision issues during the fracture assembly process and enhance the results. Our framework exhibits better performance on both PartNet and Breaking Bad datasets compared to existing state-of-the-art frameworks. Extensive experiments and quantitative comparisons demonstrate the effectiveness of our proposed framework, which features linear computational complexity, enhanced abstraction, and improved generalization. Our code is publicly available at https://github.com/Ruiyuan-Zhang/CCS.
Authors: Junru Lin, Asen Nachkov, Songyou Peng, Luc Van Gool, Danda Pani Paudel
The opacity of rigid 3D scenes with opaque surfaces is considered to be of a binary type. However, we observed that this property is not followed by the existing RGB-only NeRF-SLAM. Therefore, we are motivated to introduce this prior into the RGB-only NeRF-SLAM pipeline. Unfortunately, the optimization through the volumetric rendering function does not facilitate easy integration of the desired prior. Instead, we observed that the opacity of ternary-type (TT) is well supported. In this work, we study why ternary-type opacity is well-suited and desired for the task at hand. In particular, we provide theoretical insights into the process of jointly optimizing radiance and opacity through the volumetric rendering process. Through exhaustive experiments on benchmark datasets, we validate our claim and provide insights into the optimization process, which we believe will unleash the potential of RGB-only NeRF-SLAM. To foster this line of research, we also propose a simple yet novel visual odometry scheme that uses a hybrid combination of volumetric and warping-based image renderings. More specifically, the proposed hybrid odometry (HO) additionally uses image warping-based coarse odometry, leading up to an order of magnitude final speed-up. Furthermore, we show that the proposed TT and HO well complement each other, offering state-of-the-art results on benchmark datasets in terms of both speed and accuracy.
Authors: Tao Wu, Tie Luo, Donald C. Wunsch
The capacity to generalize to future unseen data stands as one of the utmost crucial attributes of deep neural networks. Sharpness-Aware Minimization (SAM) aims to enhance the generalizability by minimizing worst-case loss using one-step gradient ascent as an approximation. However, as training progresses, the non-linearity of the loss landscape increases, rendering one-step gradient ascent less effective. On the other hand, multi-step gradient ascent will incur higher training cost. In this paper, we introduce a normalized Hessian trace to accurately measure the curvature of loss landscape on {\em both} training and test sets. In particular, to counter excessive non-linearity of loss landscape, we propose Curvature Regularized SAM (CR-SAM), integrating the normalized Hessian trace as a SAM regularizer. Additionally, we present an efficient way to compute the trace via finite differences with parallelism. Our theoretical analysis based on PAC-Bayes bounds establishes the regularizer's efficacy in reducing generalization error. Empirical evaluation on CIFAR and ImageNet datasets shows that CR-SAM consistently enhances classification performance for ResNet and Vision Transformer (ViT) models across various datasets. Our code is available at https://github.com/TrustAIoT/CR-SAM.
Authors: Hayk Manukyan, Andranik Sargsyan, Barsegh Atanyan, Zhangyang Wang, Shant Navasardyan, Humphrey Shi
Recent progress in text-guided image inpainting, based on the unprecedented success of text-to-image diffusion models, has led to exceptionally realistic and visually plausible results. However, there is still significant potential for improvement in current text-to-image inpainting models, particularly in better aligning the inpainted area with user prompts and performing high-resolution inpainting. Therefore, in this paper we introduce HD-Painter, a completely training-free approach that accurately follows to prompts and coherently scales to high-resolution image inpainting. To this end, we design the Prompt-Aware Introverted Attention (PAIntA) layer enhancing self-attention scores by prompt information and resulting in better text alignment generations. To further improve the prompt coherence we introduce the Reweighting Attention Score Guidance (RASG) mechanism seamlessly integrating a post-hoc sampling strategy into general form of DDIM to prevent out-of-distribution latent shifts. Moreover, HD-Painter allows extension to larger scales by introducing a specialized super-resolution technique customized for inpainting, enabling the completion of missing regions in images of up to 2K resolution. Our experiments demonstrate that HD-Painter surpasses existing state-of-the-art approaches qualitatively and quantitatively, achieving an impressive generation accuracy improvement of 61.4% vs 51.9%. We will make the codes publicly available at: https://github.com/Picsart-AI-Research/HD-Painter
Authors: Penghao Wu, Saining Xie
When we look around and perform complex tasks, how we see and selectively process what we see is crucial. However, the lack of this visual search mechanism in current multimodal LLMs (MLLMs) hinders their ability to focus on important visual details, especially when handling high-resolution and visually crowded images. To address this, we introduce V*, an LLM-guided visual search mechanism that employs the world knowledge in LLMs for efficient visual querying. When combined with an MLLM, this mechanism enhances collaborative reasoning, contextual understanding, and precise targeting of specific visual elements. This integration results in a new MLLM meta-architecture, named Show, sEArch, and TelL (SEAL). We further create V*Bench, a benchmark specifically designed to evaluate MLLMs in their ability to process high-resolution images and focus on visual details. Our study highlights the necessity of incorporating visual search capabilities into multimodal systems. The code is available https://github.com/penghao-wu/vstar.
Authors: Seungjun An, Seonghoon Park, Gyeongnyeon Kim, Jeongyeol Baek, Byeongwon Lee, Seungryong Kim
With the increasing importance of video data in real-world applications, there is a rising need for efficient object detection methods that utilize temporal information. While existing video object detection (VOD) techniques employ various strategies to address this challenge, they typically depend on locally adjacent frames or randomly sampled images within a clip. Although recent Transformer-based VOD methods have shown promising results, their reliance on multiple inputs and additional network complexity to incorporate temporal information limits their practical applicability. In this paper, we propose a novel approach to single image object detection, called Context Enhanced TRansformer (CETR), by incorporating temporal context into DETR using a newly designed memory module. To efficiently store temporal information, we construct a class-wise memory that collects contextual information across data. Additionally, we present a classification-based sampling technique to selectively utilize the relevant memory for the current image. In the testing, We introduce a test-time memory adaptation method that updates individual memory functions by considering the test distribution. Experiments with CityCam and ImageNet VID datasets exhibit the efficiency of the framework on various video systems. The project page and code will be made available at: https://ku-cvlab.github.io/CETR.
Authors: James Gunn, Zygmunt Lenyk, Anuj Sharma, Andrea Donati, Alexandru Buburuzan, John Redford, Romain Mueller
Combining complementary sensor modalities is crucial to providing robust perception for safety-critical robotics applications such as autonomous driving (AD). Recent state-of-the-art camera-lidar fusion methods for AD rely on monocular depth estimation which is a notoriously difficult task compared to using depth information from the lidar directly. Here, we find that this approach does not leverage depth as expected and show that naively improving depth estimation does not lead to improvements in object detection performance and that, strikingly, removing depth estimation altogether does not degrade object detection performance. This suggests that relying on monocular depth could be an unnecessary architectural bottleneck during camera-lidar fusion. In this work, we introduce a novel fusion method that bypasses monocular depth estimation altogether and instead selects and fuses camera and lidar features in a bird's-eye-view grid using a simple attention mechanism. We show that our model can modulate its use of camera features based on the availability of lidar features and that it yields better 3D object detection on the nuScenes dataset than baselines relying on monocular depth estimation.
Authors: Riccardo Scodellaro, Ajinkya Kulkarni, Frauke Alves, Matthias Schröter
The recent successes in analyzing images with deep neural networks are almost exclusively achieved with Convolutional Neural Networks (CNNs). The training of these CNNs, and in fact of all deep neural network architectures, uses the backpropagation algorithm where the output of the network is compared with the desired result and the difference is then used to tune the weights of the network towards the desired outcome. In a 2022 preprint, Geoffrey Hinton suggested an alternative way of training which passes the desired results together with the images at the input of the network. This so called Forward Forward (FF) algorithm has up to now only been used in fully connected networks. In this paper, we show how the FF paradigm can be extended to CNNs. Our FF-trained CNN, featuring a novel spatially-extended labeling technique, achieves a classification accuracy of 99.0% on the MNIST hand-written digits dataset. We show how different hyperparameters affect the performance of the proposed algorithm and compare the results with CNN trained with the standard backpropagation approach. Furthermore, we use Class Activation Maps to investigate which type of features are learnt by the FF algorithm.