Authors: Mingquan Lin, Song Wang, Ying Ding, Lihui Zhao, Fei Wang, Yifan Peng
Background: The predictive Intensive Care Unit (ICU) scoring system plays an important role in ICU management because it predicts important outcomes, especially mortality. Many scoring systems have been developed and used in the ICU. These scoring systems are primarily based on the structured clinical data in the electronic health record (EHR), which may suffer the loss of important clinical information in the narratives and images. Methods: In this work, we build a deep learning based survival prediction model with multi-modality data to predict ICU mortality. Four sets of features are investigated: (1) physiological measurements of Simplified Acute Physiology Score (SAPS) II, (2) common thorax diseases pre-defined by radiologists, (3) BERT-based text representations, and (4) chest X-ray image features. We use the Medical Information Mart for Intensive Care IV (MIMIC-IV) dataset to evaluate the proposed model. Results: Our model achieves the average C-index of 0.7829 (95% confidence interval, 0.7620-0.8038), which substantially exceeds that of the baseline with SAPS-II features (0.7470 (0.7263-0.7676)). Ablation studies further demonstrate the contributions of pre-defined labels (2.00%), text features (2.44%), and image features (2.82%).
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: Lei Ma, Jincong Han, Zhaoxin Wang, Dian Zhang
Large-scale multimodal language models (LMMs) have achieved remarkable success in general domains. However, the exploration of diagnostic language models based on multimodal cephalometric medical data remains limited. In this paper, we propose a novel multimodal cephalometric analysis and diagnostic dialogue model. Firstly, a multimodal orthodontic medical dataset is constructed, comprising cephalometric images and doctor-patient dialogue data, with automatic analysis of cephalometric landmarks using U-net and generation of diagnostic reports. Then, the cephalometric dataset and generated diagnostic reports are separately fine-tuned on Minigpt-4 and VisualGLM. Results demonstrate that the CephGPT-4 model exhibits excellent performance and has the potential to revolutionize orthodontic measurement and diagnostic applications. These innovations hold revolutionary application potential in the field of orthodontics.
Authors: Cristian Camilo Pulgarín-Ospina, Rocío del Amor, Adrián Colomera, Julio Silva-Rodríguez, Valery Naranjo
Digital pathology has become a standard in the pathology workflow due to its many benefits. These include the level of detail of the whole slide images generated and the potential immediate sharing of cases between hospitals. Recent advances in deep learning-based methods for image analysis make them of potential aid in digital pathology. However, a major limitation in developing computer-aided diagnostic systems for pathology is the lack of an intuitive and open web application for data annotation. This paper proposes a web service that efficiently provides a tool to visualize and annotate digitized histological images. In addition, to show and validate the tool, in this paper we include a use case centered on the diagnosis of spindle cell skin neoplasm for multiple annotators. A usability study of the tool is also presented, showing the feasibility of the developed tool.
Authors: Cedric Walker, Tasneem Talawalla, Robert Toth, Akhil Ambekar, Kien Rea, Oswin Chamian, Fan Fan, Sabina Berezowska, Sven Rottenberg, Anant Madabhushi, Marie Maillard, Laura Barisoni, Hugo Mark Horlings, Andrew Janowczyk
The discovery of patterns associated with diagnosis, prognosis, and therapy response in digital pathology images often requires intractable labeling of large quantities of histological objects. Here we release an open-source labeling tool, PatchSorter, which integrates deep learning with an intuitive web interface. Using >100,000 objects, we demonstrate a >7x improvement in labels per second over unaided labeling, with minimal impact on labeling accuracy, thus enabling high-throughput labeling of large datasets.
Authors: Mariana-Iuliana Georgescu
Pathological anomalies exhibit diverse appearances in medical imaging, making it difficult to collect and annotate a representative amount of data required to train deep learning models in a supervised setting. Therefore, in this work, we tackle anomaly detection in medical images training our framework using only healthy samples. We propose to use the Masked Autoencoder model to learn the structure of the normal samples, then train an anomaly classifier on top of the difference between the original image and the reconstruction provided by the masked autoencoder. We train the anomaly classifier in a supervised manner using as negative samples the reconstruction of the healthy scans, while as positive samples, we use pseudo-abnormal scans obtained via our novel pseudo-abnormal module. The pseudo-abnormal module alters the reconstruction of the normal samples by changing the intensity of several regions. We conduct experiments on two medical image data sets, namely BRATS2020 and LUNA16 and compare our method with four state-of-the-art anomaly detection frameworks, namely AST, RD4AD, AnoVAEGAN and f-AnoGAN.
Authors: Chengyu Fang, Xianfeng Han
In this paper, we investigate the problem of automatically controllable artistic character line drawing generation from photographs by proposing a Vector Flow Aware and Line Controllable Image-to-Image Translation architecture, which can be viewed as an appealing intersection between Artificial Intelligence and Arts. Specifically, we first present an Image-to-Flow network (I2FNet) to efficiently and robustly create the vector flow field in a learning-based manner, which can provide a direction guide for drawing lines. Then, we introduce our well-designed Double Flow Generator (DFG) framework to fuse features from learned vector flow and input image flow guaranteeing the spatial coherence of lines. Meanwhile, in order to allow for controllable character line drawing generation, we integrate a Line Control Matrix (LCM) into DFG and train a Line Control Regressor (LCR) to synthesize drawings with different styles by elaborately controlling the level of details, such as thickness, smoothness, and continuity, of lines. Finally, we design a Fourier Transformation Loss to further constrain the character line generation from the frequency domain view of the point. Quantitative and qualitative experiments demonstrate that our approach can obtain superior performance in producing high-resolution character line-drawing images with perceptually realistic characteristics.
Authors: Marc Demoustier, Yue Zhang, Venkatesh Narasimha Murthy, Florin C. Ghesu, Dorin Comaniciu
Device tracking is an important prerequisite for guidance during endovascular procedures. Especially during cardiac interventions, detection and tracking of guiding the catheter tip in 2D fluoroscopic images is important for applications such as mapping vessels from angiography (high dose with contrast) to fluoroscopy (low dose without contrast). Tracking the catheter tip poses different challenges: the tip can be occluded by contrast during angiography or interventional devices; and it is always in continuous movement due to the cardiac and respiratory motions. To overcome these challenges, we propose ConTrack, a transformer-based network that uses both spatial and temporal contextual information for accurate device detection and tracking in both X-ray fluoroscopy and angiography. The spatial information comes from the template frames and the segmentation module: the template frames define the surroundings of the device, whereas the segmentation module detects the entire device to bring more context for the tip prediction. Using multiple templates makes the model more robust to the change in appearance of the device when it is occluded by the contrast agent. The flow information computed on the segmented catheter mask between the current and the previous frame helps in further refining the prediction by compensating for the respiratory and cardiac motions. The experiments show that our method achieves 45% or higher accuracy in detection and tracking when compared to state-of-the-art tracking models.
Authors: Dibya Nath, G. M. Shahariar
Gastrointestinal diseases pose significant healthcare chall-enges as they manifest in diverse ways and can lead to potential complications. Ensuring precise and timely classification of these diseases is pivotal in guiding treatment choices and enhancing patient outcomes. This paper introduces a novel approach on classifying gastrointestinal diseases by leveraging cost-sensitive pre-trained deep convolutional neural network (CNN) architectures with supervised contrastive learning. Our approach enables the network to learn representations that capture vital disease-related features, while also considering the relationships of similarity between samples. To tackle the challenges posed by imbalanced datasets and the cost-sensitive nature of misclassification errors in healthcare, we incorporate cost-sensitive learning. By assigning distinct costs to misclassifications based on the disease class, we prioritize accurate classification of critical conditions. Furthermore, we enhance the interpretability of our model by integrating gradient-based techniques from explainable artificial intelligence (AI). This inclusion provides valuable insights into the decision-making process of the network, aiding in understanding the features that contribute to disease classification. To assess the effectiveness of our proposed approach, we perform extensive experiments on a comprehensive gastrointestinal disease dataset, such as the Hyper-Kvasir dataset. Through thorough comparisons with existing works, we demonstrate the strong classification accuracy, robustness and interpretability of our model. We have made the implementation of our proposed approach publicly available at https://github.com/dibya404/Gastrointestinal-Disease-Classification-through-Explainable-and-Cost-Sensitive-DNN-with-SCL
Authors: Yeti Z. Gurbuz, A. Aydin Alatan
Global average pooling (GAP) is a popular component in deep metric learning (DML) for aggregating features. Its effectiveness is often attributed to treating each feature vector as a distinct semantic entity and GAP as a combination of them. Albeit substantiated, such an explanation's algorithmic implications to learn generalizable entities to represent unseen classes, a crucial DML goal, remain unclear. To address this, we formulate GAP as a convex combination of learnable prototypes. We then show that the prototype learning can be expressed as a recursive process fitting a linear predictor to a batch of samples. Building on that perspective, we consider two batches of disjoint classes at each iteration and regularize the learning by expressing the samples of a batch with the prototypes that are fitted to the other batch. We validate our approach on 4 popular DML benchmarks.
Authors: Nikita Karaev, Ignacio Rocco, Benjamin Graham, Natalia Neverova, Andrea Vedaldi, Christian Rupprecht
Methods for video motion prediction either estimate jointly the instantaneous motion of all points in a given video frame using optical flow or independently track the motion of individual points throughout the video. The latter is true even for powerful deep-learning methods that can track points through occlusions. Tracking points individually ignores the strong correlation that can exist between the points, for instance, because they belong to the same physical object, potentially harming performance. In this paper, we thus propose CoTracker, an architecture that jointly tracks multiple points throughout an entire video. This architecture combines several ideas from the optical flow and tracking literature in a new, flexible and powerful design. It is based on a transformer network that models the correlation of different points in time via specialised attention layers. The transformer iteratively updates an estimate of several trajectories. It can be applied in a sliding-window manner to very long videos, for which we engineer an unrolled training loop. It can track from one to several points jointly and supports adding new points to track at any time. The result is a flexible and powerful tracking algorithm that outperforms state-of-the-art methods in almost all benchmarks.
Authors: Chenyu Zhang, Zhaozheng Yin, Ruwen Qin
Efficiently monitoring the condition of civil infrastructures necessitates automating the structural condition assessment in visual inspection. This paper proposes an Attention-enhanced Co-interactive Fusion Network (ACF-Net) for automatic structural condition assessment in visual bridge inspection. The ACF-Net can simultaneously parse structural elements and segment surface defects on the elements in inspection images. It integrates two task-specific relearning subnets to extract task-specific features from an overall feature embedding and a co-interactive feature fusion module to capture the spatial correlation and facilitate information sharing between tasks. Experimental results demonstrate that the proposed ACF-Net outperforms the current state-of-the-art approaches, achieving promising performance with 92.11% mIoU for element parsing and 87.16% mIoU for corrosion segmentation on the new benchmark dataset Steel Bridge Condition Inspection Visual (SBCIV) testing set. An ablation study reveals the strengths of ACF-Net, and a case study showcases its capability to automate structural condition assessment. The code will be open-source after acceptance.
Authors: Donghua Wang, Wen Yao, Tingsong Jiang, Chao Li, Xiaoqian Chen
Physical adversarial attacks against deep neural networks (DNNs) have recently gained increasing attention. The current mainstream physical attacks use printed adversarial patches or camouflage to alter the appearance of the target object. However, these approaches generate conspicuous adversarial patterns that show poor stealthiness. Another physical deployable attack is the optical attack, featuring stealthiness while exhibiting weakly in the daytime with sunlight. In this paper, we propose a novel Reflected Light Attack (RFLA), featuring effective and stealthy in both the digital and physical world, which is implemented by placing the color transparent plastic sheet and a paper cut of a specific shape in front of the mirror to create different colored geometries on the target object. To achieve these goals, we devise a general framework based on the circle to model the reflected light on the target object. Specifically, we optimize a circle (composed of a coordinate and radius) to carry various geometrical shapes determined by the optimized angle. The fill color of the geometry shape and its corresponding transparency are also optimized. We extensively evaluate the effectiveness of RFLA on different datasets and models. Experiment results suggest that the proposed method achieves over 99% success rate on different datasets and models in the digital world. Additionally, we verify the effectiveness of the proposed method in different physical environments by using sunlight or a flashlight.
Authors: Ma Siliang, Xu Yong
Bounding box regression (BBR) has been widely used in object detection and instance segmentation, which is an important step in object localization. However, most of the existing loss functions for bounding box regression cannot be optimized when the predicted box has the same aspect ratio as the groundtruth box, but the width and height values are exactly different. In order to tackle the issues mentioned above, we fully explore the geometric features of horizontal rectangle and propose a novel bounding box similarity comparison metric MPDIoU based on minimum point distance, which contains all of the relevant factors considered in the existing loss functions, namely overlapping or non-overlapping area, central points distance, and deviation of width and height, while simplifying the calculation process. On this basis, we propose a bounding box regression loss function based on MPDIoU, called LMPDIoU . Experimental results show that the MPDIoU loss function is applied to state-of-the-art instance segmentation (e.g., YOLACT) and object detection (e.g., YOLOv7) model trained on PASCAL VOC, MS COCO, and IIIT5k outperforms existing loss functions.
Authors: Jiahui Huang, Leonid Sigal, Kwang Moo Yi, Oliver Wang, Joon-Young Lee
We present Interactive Neural Video Editing (INVE), a real-time video editing solution, which can assist the video editing process by consistently propagating sparse frame edits to the entire video clip. Our method is inspired by the recent work on Layered Neural Atlas (LNA). LNA, however, suffers from two major drawbacks: (1) the method is too slow for interactive editing, and (2) it offers insufficient support for some editing use cases, including direct frame editing and rigid texture tracking. To address these challenges we leverage and adopt highly efficient network architectures, powered by hash-grids encoding, to substantially improve processing speed. In addition, we learn bi-directional functions between image-atlas and introduce vectorized editing, which collectively enables a much greater variety of edits in both the atlas and the frames directly. Compared to LNA, our INVE reduces the learning and inference time by a factor of 5, and supports various video editing operations that LNA cannot. We showcase the superiority of INVE over LNA in interactive video editing through a comprehensive quantitative and qualitative analysis, highlighting its numerous advantages and improved performance. For video results, please see https://gabriel-huang.github.io/inve/
Authors: Jingyi Xu, Hieu Le, Dimitris Samaras
Class-agnostic counting (CAC) has numerous potential applications across various domains. The goal is to count objects of an arbitrary category during testing, based on only a few annotated exemplars. In this paper, we point out that the task of counting objects of interest when there are multiple object classes in the image (namely, multi-class object counting) is particularly challenging for current object counting models. They often greedily count every object regardless of the exemplars. To address this issue, we propose localizing the area containing the objects of interest via an exemplar-based segmentation model before counting them. The key challenge here is the lack of segmentation supervision to train this model. To this end, we propose a method to obtain pseudo segmentation masks using only box exemplars and dot annotations. We show that the segmentation model trained on these pseudo-labeled masks can effectively localize objects of interest for an arbitrary multi-class image based on the exemplars. To evaluate the performance of different methods on multi-class counting, we introduce two new benchmarks, a synthetic multi-class dataset and a new test set of real images in which objects from multiple classes are present. Our proposed method shows a significant advantage over the previous CAC methods on these two benchmarks.
Authors: Ze Lu, Yalei Lv, Wenqi Wang, Pengfei Xiong
Deep generative approaches have obtained great success in image inpainting recently. However, most generative inpainting networks suffer from either over-smooth results or aliasing artifacts. The former lacks high-frequency details, while the latter lacks semantic structure. To address this issue, we propose an effective Frequency-Spatial Complementary Network (FSCN) by exploiting rich semantic information in both spatial and frequency domains. Specifically, we introduce an extra Frequency Branch and Frequency Loss on the spatial-based network to impose direct supervision on the frequency information, and propose a Frequency-Spatial Cross-Attention Block (FSCAB) to fuse multi-domain features and combine the corresponding characteristics. With our FSCAB, the inpainting network is capable of capturing frequency information and preserving visual consistency simultaneously. Extensive quantitative and qualitative experiments demonstrate that our inpainting network can effectively achieve superior results, outperforming previous state-of-the-art approaches with significantly fewer parameters and less computation cost. The code will be released soon.
Authors: Cheng Chen, Yifan Zhao, Jia Li
Learning multi-label image recognition with incomplete annotation is gaining popularity due to its superior performance and significant labor savings when compared to training with fully labeled datasets. Existing literature mainly focuses on label completion and co-occurrence learning while facing difficulties with the most common single-positive label manner. To tackle this problem, we present a semantic contrastive bootstrapping (Scob) approach to gradually recover the cross-object relationships by introducing class activation as semantic guidance. With this learning guidance, we then propose a recurrent semantic masked transformer to extract iconic object-level representations and delve into the contrastive learning problems on multi-label classification tasks. We further propose a bootstrapping framework in an Expectation-Maximization fashion that iteratively optimizes the network parameters and refines semantic guidance to alleviate possible disturbance caused by wrong semantic guidance. Extensive experimental results demonstrate that the proposed joint learning framework surpasses the state-of-the-art models by a large margin on four public multi-label image recognition benchmarks. Codes can be found at https://github.com/iCVTEAM/Scob.
Authors: Yuanshuo Cheng, Mingwen Shao, Yecong Wan, Chao Wang, Wangmeng Zuo
Existing All-In-One image restoration (IR) methods usually lack flexible modeling on various types of degradation, thus impeding the restoration performance. To achieve All-In-One IR with higher task dexterity, this work proposes an efficient Dynamic Reference Modeling paradigm (DRM-IR), which consists of task-adaptive degradation modeling and model-based image restoring. Specifically, these two subtasks are formalized as a pair of entangled reference-based maximum a posteriori (MAP) inferences, which are optimized synchronously in an unfolding-based manner. With the two cascaded subtasks, DRM-IR first dynamically models the task-specific degradation based on a reference image pair and further restores the image with the collected degradation statistics. Besides, to bridge the semantic gap between the reference and target degraded images, we further devise a Degradation Prior Transmitter (DPT) that restrains the instance-specific feature differences. DRM-IR explicitly provides superior flexibility for All-in-One IR while being interpretable. Extensive experiments on multiple benchmark datasets show that our DRM-IR achieves state-of-the-art in All-In-One IR.
Authors: Abhinandan Kumar Pun, Mohammed Javed, David S. Doermann
The problem of change detection in images finds application in different domains like diagnosis of diseases in the medical field, detecting growth patterns of cities through remote sensing, and finding changes in legal documents and contracts. However, this paper presents a survey on core techniques and rules to detect changes in different versions of a document image. Our discussions on change detection focus on two categories -- content-based and layout-based. The content-based techniques intelligently extract and analyze the image contents (text or non-text) to show the possible differences, whereas the layout-based techniques use structural information to predict document changes. We also summarize the existing datasets and evaluation metrics used in change detection experiments. The shortcomings and challenges the existing methods face are reported, along with some pointers for future research work.
Authors: Meng Ye, Dong Yang, Mikael Kanski, Leon Axel, Dimitris Metaxas
We propose a novel neural deformable model (NDM) targeting at the reconstruction and modeling of 3D bi-ventricular shape of the heart from 2D sparse cardiac magnetic resonance (CMR) imaging data. We model the bi-ventricular shape using blended deformable superquadrics, which are parameterized by a set of geometric parameter functions and are capable of deforming globally and locally. While global geometric parameter functions and deformations capture gross shape features from visual data, local deformations, parameterized as neural diffeomorphic point flows, can be learned to recover the detailed heart shape.Different from iterative optimization methods used in conventional deformable model formulations, NDMs can be trained to learn such geometric parameter functions, global and local deformations from a shape distribution manifold. Our NDM can learn to densify a sparse cardiac point cloud with arbitrary scales and generate high-quality triangular meshes automatically. It also enables the implicit learning of dense correspondences among different heart shape instances for accurate cardiac shape registration. Furthermore, the parameters of NDM are intuitive, and can be used by a physician without sophisticated post-processing. Experimental results on a large CMR dataset demonstrate the improved performance of NDM over conventional methods.
Authors: Lei Pan, Wuyang Luan, Yuan Zheng, Qiang Fu, Junhui Li
Most existing 3D instance segmentation methods are derived from 3D semantic segmentation models. However, these indirect approaches suffer from certain limitations. They fail to fully leverage global and local semantic information for accurate prediction, which hampers the overall performance of the 3D instance segmentation framework. To address these issues, this paper presents PSGformer, a novel 3D instance segmentation network. PSGformer incorporates two key advancements to enhance the performance of 3D instance segmentation. Firstly, we propose a Multi-Level Semantic Aggregation Module, which effectively captures scene features by employing foreground point filtering and multi-radius aggregation. This module enables the acquisition of more detailed semantic information from global and local perspectives. Secondly, PSGformer introduces a Parallel Feature Fusion Transformer Module that independently processes super-point features and aggregated features using transformers. The model achieves a more comprehensive feature representation by the features which connect global and local features. We conducted extensive experiments on the ScanNetv2 dataset. Notably, PSGformer exceeds compared state-of-the-art methods by 2.2% on ScanNetv2 hidden test set in terms of mAP. Our code and models will be publicly released.
Authors: Yufei Wang, Yi Yu, Wenhan Yang, Lanqing Guo, Lap-Pui Chau, Alex C. Kot, Bihan Wen
Previous raw image-based low-light image enhancement methods predominantly relied on feed-forward neural networks to learn deterministic mappings from low-light to normally-exposed images. However, they failed to capture critical distribution information, leading to visually undesirable results. This work addresses the issue by seamlessly integrating a diffusion model with a physics-based exposure model. Different from a vanilla diffusion model that has to perform Gaussian denoising, with the injected physics-based exposure model, our restoration process can directly start from a noisy image instead of pure noise. As such, our method obtains significantly improved performance and reduced inference time compared with vanilla diffusion models. To make full use of the advantages of different intermediate steps, we further propose an adaptive residual layer that effectively screens out the side-effect in the iterative refinement when the intermediate results have been already well-exposed. The proposed framework can work with both real-paired datasets, SOTA noise models, and different backbone networks. Note that, the proposed framework is compatible with real-paired datasets, real/synthetic noise models, and different backbone networks. We evaluate the proposed method on various public benchmarks, achieving promising results with consistent improvements using different exposure models and backbones. Besides, the proposed method achieves better generalization capacity for unseen amplifying ratios and better performance than a larger feedforward neural model when few parameters are adopted.
Authors: Guandong Li, Mengxia Ye
Deep neural networks have faced many problems in hyperspectral image classification, including the ineffective utilization of spectral-spatial joint information and the problems of gradient vanishing and overfitting that arise with increasing depth. In order to accelerate the deployment of models on edge devices with strict latency requirements and limited computing power, this paper proposes a learnable group convolution network (LGCNet) based on an improved 3D-DenseNet model and a lightweight model design. The LGCNet module improves the shortcomings of group convolution by introducing a dynamic learning method for the input channels and convolution kernel grouping, enabling flexible grouping structures and generating better representation ability. Through the overall loss and gradient of the backpropagation network, the 3D group convolution is dynamically determined and updated in an end-to-end manner. The learnable number of channels and corresponding grouping can capture different complementary visual features of input images, allowing the CNN to learn richer feature representations. When extracting high-dimensional and redundant hyperspectral data, the 3D convolution kernels also contain a large amount of redundant information. The LGC module allows the 3D-DenseNet to choose channel information with more semantic features, and is very efficient, making it suitable for embedding in any deep neural network for acceleration and efficiency improvements. LGC enables the 3D-CNN to achieve sufficient feature extraction while also meeting speed and computing requirements. Furthermore, LGCNet has achieved progress in inference speed and accuracy, and outperforms mainstream hyperspectral image classification methods on the Indian Pines, Pavia University, and KSC datasets.
Authors: Kensuke Mukai, Takao Yamanaka
Zero padding is often used in convolutional neural networks to prevent the feature map size from decreasing with each layer. However, recent studies have shown that zero padding promotes encoding of absolute positional information, which may adversely affect the performance of some tasks. In this work, a novel padding method called Peripheral Prediction Padding (PP-Pad) method is proposed, which enables end-to-end training of padding values suitable for each task instead of zero padding. Moreover, novel metrics to quantitatively evaluate the translation invariance of the model are presented. By evaluating with these metrics, it was confirmed that the proposed method achieved higher accuracy and translation invariance than the previous methods in a semantic segmentation task.
Authors: Hinata Aoki, Takao Yamanaka
Neural Radiance Fields (NeRF) has been applied to various tasks related to representations of 3D scenes. Most studies based on NeRF have focused on a small object, while a few studies have tried to reconstruct large-scale scenes although these methods tend to require large computational cost. For the application of NeRF to large-scale scenes, a method based on NeRF is proposed in this paper to effectively use height data which can be obtained from GIS (Geographic Information System). For this purpose, the scene space was divided into multiple objects and a background using the height data to represent them with separate neural networks. In addition, an adaptive sampling method is also proposed by using the height data. As a result, the accuracy of image rendering was improved with faster training speed.
Authors: Alzayat Saleh, Md Mehedi Hasan, Herman W Raadsma, Mehar S Khatkar, Dean R Jerry, Mostafa Rahimi Azghadi
Accurate weight estimation and morphometric analyses are useful in aquaculture for optimizing feeding, predicting harvest yields, identifying desirable traits for selective breeding, grading processes, and monitoring the health status of production animals. However, the collection of phenotypic data through traditional manual approaches at industrial scales and in real-time is time-consuming, labour-intensive, and prone to errors. Digital imaging of individuals and subsequent training of prediction models using Deep Learning (DL) has the potential to rapidly and accurately acquire phenotypic data from aquaculture species. In this study, we applied a novel DL approach to automate weight estimation and morphometric analysis using the black tiger prawn (Penaeus monodon) as a model crustacean. The DL approach comprises two main components: a feature extraction module that efficiently combines low-level and high-level features using the Kronecker product operation; followed by a landmark localization module that then uses these features to predict the coordinates of key morphological points (landmarks) on the prawn body. Once these landmarks were extracted, weight was estimated using a weight regression module based on the extracted landmarks using a fully connected network. For morphometric analyses, we utilized the detected landmarks to derive five important prawn traits. Principal Component Analysis (PCA) was also used to identify landmark-derived distances, which were found to be highly correlated with shape features such as body length, and width. We evaluated our approach on a large dataset of 8164 images of the Black tiger prawn (Penaeus monodon) collected from Australian farms. Our experimental results demonstrate that the novel DL approach outperforms existing DL methods in terms of accuracy, robustness, and efficiency.
Authors: Yi-Syuan Chen, Yun-Zhu Song, Cheng Yu Yeo, Bei Liu, Jianlong Fu, Hong-Han Shuai
Large Pre-trained Transformers exhibit an intriguing capacity for in-context learning. Without gradient updates, these models can rapidly construct new predictors from demonstrations presented in the inputs. Recent works promote this ability in the vision-language domain by incorporating visual information into large language models that can already make in-context predictions. However, these methods could inherit issues in the language domain, such as template sensitivity and hallucination. Also, the scale of these language models raises a significant demand for computations, making learning and operating these models resource-intensive. To this end, we raise a question: ``How can we enable in-context learning for general models without being constrained on large language models?". To answer it, we propose a succinct and general framework, Self-supervised IN-Context learning (SINC), that introduces a meta-model to learn on self-supervised prompts consisting of tailored demonstrations. The learned models can be transferred to downstream tasks for making in-context predictions on-the-fly. Extensive experiments show that SINC outperforms gradient-based methods in various vision-language tasks under few-shot settings. Furthermore, the designs of SINC help us investigate the benefits of in-context learning across different tasks, and the analysis further reveals the essential components for the emergence of in-context learning in the vision-language domain.
Authors: Wing-Yin Yu, Lai-Man Po, Ray Cheung, Yuzhi Zhao, Yu Xue, Kun Li
Video-based human pose transfer is a video-to-video generation task that animates a plain source human image based on a series of target human poses. Considering the difficulties in transferring highly structural patterns on the garments and discontinuous poses, existing methods often generate unsatisfactory results such as distorted textures and flickering artifacts. To address these issues, we propose a novel Deformable Motion Modulation (DMM) that utilizes geometric kernel offset with adaptive weight modulation to simultaneously perform feature alignment and style transfer. Different from normal style modulation used in style transfer, the proposed modulation mechanism adaptively reconstructs smoothed frames from style codes according to the object shape through an irregular receptive field of view. To enhance the spatio-temporal consistency, we leverage bidirectional propagation to extract the hidden motion information from a warped image sequence generated by noisy poses. The proposed feature propagation significantly enhances the motion prediction ability by forward and backward propagation. Both quantitative and qualitative experimental results demonstrate superiority over the state-of-the-arts in terms of image fidelity and visual continuity. The source code is publicly available at github.com/rocketappslab/bdmm.
Authors: Ruiping Liu, Jiaming Zhang, Kunyu Peng, Junwei Zheng, Ke Cao, Yufan Chen, Kailun Yang, Rainer Stiefelhagen
Grounded Situation Recognition (GSR) is capable of recognizing and interpreting visual scenes in a contextually intuitive way, yielding salient activities (verbs) and the involved entities (roles) depicted in images. In this work, we focus on the application of GSR in assisting people with visual impairments (PVI). However, precise localization information of detected objects is often required to navigate their surroundings confidently and make informed decisions. For the first time, we propose an Open Scene Understanding (OpenSU) system that aims to generate pixel-wise dense segmentation masks of involved entities instead of bounding boxes. Specifically, we build our OpenSU system on top of GSR by additionally adopting an efficient Segment Anything Model (SAM). Furthermore, to enhance the feature extraction and interaction between the encoder-decoder structure, we construct our OpenSU system using a solid pure transformer backbone to improve the performance of GSR. In order to accelerate the convergence, we replace all the activation functions within the GSR decoders with GELU, thereby reducing the training duration. In quantitative analysis, our model achieves state-of-the-art performance on the SWiG dataset. Moreover, through field testing on dedicated assistive technology datasets and application demonstrations, the proposed OpenSU system can be used to enhance scene understanding and facilitate the independent mobility of people with visual impairments. Our code will be available at https://github.com/RuipingL/OpenSU.
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: Sarosij Bose, Saikat Sarkar, Amlan Chakrabarti
Classifying player actions from soccer videos is a challenging problem, which has become increasingly important in sports analytics over the years. Most state-of-the-art methods employ highly complex offline networks, which makes it difficult to deploy such models in resource constrained scenarios. Here, in this paper we propose a novel end-to-end knowledge distillation based transfer learning network pre-trained on the Kinetics400 dataset and then perform extensive analysis on the learned framework by introducing a unique loss parameterization. We also introduce a new dataset named SoccerDB1 containing 448 videos and consisting of 4 diverse classes each of players playing soccer. Furthermore, we introduce an unique loss parameter that help us linearly weigh the extent to which the predictions of each network are utilized. Finally, we also perform a thorough performance study using various changed hyperparameters. We also benchmark the first classification results on the new SoccerDB1 dataset obtaining 67.20% validation accuracy. Apart from outperforming prior arts significantly, our model also generalizes to new datasets easily. The dataset has been made publicly available at: https://bit.ly/soccerdb1
Authors: Zhizhong Huang, Siteng Ma, Junping Zhang, Hongming Shan
Recent works for face editing usually manipulate the latent space of StyleGAN via the linear semantic directions. However, they usually suffer from the entanglement of facial attributes, need to tune the optimal editing strength, and are limited to binary attributes with strong supervision signals. This paper proposes a novel adaptive nonlinear latent transformation for disentangled and conditional face editing, termed AdaTrans. Specifically, our AdaTrans divides the manipulation process into several finer steps; i.e., the direction and size at each step are conditioned on both the facial attributes and the latent codes. In this way, AdaTrans describes an adaptive nonlinear transformation trajectory to manipulate the faces into target attributes while keeping other attributes unchanged. Then, AdaTrans leverages a predefined density model to constrain the learned trajectory in the distribution of latent codes by maximizing the likelihood of transformed latent code. Moreover, we also propose a disentangled learning strategy under a mutual information framework to eliminate the entanglement among attributes, which can further relax the need for labeled data. Consequently, AdaTrans enables a controllable face editing with the advantages of disentanglement, flexibility with non-binary attributes, and high fidelity. Extensive experimental results on various facial attributes demonstrate the qualitative and quantitative effectiveness of the proposed AdaTrans over existing state-of-the-art methods, especially in the most challenging scenarios with a large age gap and few labeled examples. The source code is available at https://github.com/Hzzone/AdaTrans.
Authors: Tianyu Guo, Mengyuan Liu, Hong Liu, Wenhao Li, Jingwen Guo, Tao Wang, Yidi Li
Considering the instance-level discriminative ability, contrastive learning methods, including MoCo and SimCLR, have been adapted from the original image representation learning task to solve the self-supervised skeleton-based action recognition task. These methods usually use multiple data streams (i.e., joint, motion, and bone) for ensemble learning, meanwhile, how to construct a discriminative feature space within a single stream and effectively aggregate the information from multiple streams remains an open problem. To this end, we first apply a new contrastive learning method called BYOL to learn from skeleton data and formulate SkeletonBYOL as a simple yet effective baseline for self-supervised skeleton-based action recognition. Inspired by SkeletonBYOL, we further present a joint Adversarial and Collaborative Learning (ACL) framework, which combines Cross-Model Adversarial Learning (CMAL) and Cross-Stream Collaborative Learning (CSCL). Specifically, CMAL learns single-stream representation by cross-model adversarial loss to obtain more discriminative features. To aggregate and interact with multi-stream information, CSCL is designed by generating similarity pseudo label of ensemble learning as supervision and guiding feature generation for individual streams. Exhaustive experiments on three datasets verify the complementary properties between CMAL and CSCL and also verify that our method can perform favorably against state-of-the-art methods using various evaluation protocols. Our code and models are publicly available at \url{https://github.com/Levigty/ACL}.
Authors: Junyu Li, Han Huang, Dong Ni, Wufeng Xue, Dongmei Zhu, Jun Cheng
Early diagnosis of renal cancer can greatly improve the survival rate of patients. Contrast-enhanced ultrasound (CEUS) is a cost-effective and non-invasive imaging technique and has become more and more frequently used for renal tumor diagnosis. However, the classification of benign and malignant renal tumors can still be very challenging due to the highly heterogeneous appearance of cancer and imaging artifacts. Our aim is to detect and classify renal tumors by integrating B-mode and CEUS-mode ultrasound videos. To this end, we propose a novel multi-modal ultrasound video fusion network that can effectively perform multi-modal feature fusion and video classification for renal tumor diagnosis. The attention-based multi-modal fusion module uses cross-attention and self-attention to extract modality-invariant features and modality-specific features in parallel. In addition, we design an object-level temporal aggregation (OTA) module that can automatically filter low-quality features and efficiently integrate temporal information from multiple frames to improve the accuracy of tumor diagnosis. Experimental results on a multicenter dataset show that the proposed framework outperforms the single-modal models and the competing methods. Furthermore, our OTA module achieves higher classification accuracy than the frame-level predictions. Our code is available at \url{https://github.com/JeunyuLi/MUAF}.
Authors: Mennatullah Siam, Rezaul Karim, He Zhao, Richard Wildes
Few-shot video segmentation is the task of delineating a specific novel class in a query video using few labelled support images. Typical approaches compare support and query features while limiting comparisons to a single feature layer and thereby ignore potentially valuable information. We present a meta-learned Multiscale Memory Comparator (MMC) for few-shot video segmentation that combines information across scales within a transformer decoder. Typical multiscale transformer decoders for segmentation tasks learn a compressed representation, their queries, through information exchange across scales. Unlike previous work, we instead preserve the detailed feature maps during across scale information exchange via a multiscale memory transformer decoding to reduce confusion between the background and novel class. Integral to the approach, we investigate multiple forms of information exchange across scales in different tasks and provide insights with empirical evidence on which to use in each task. The overall comparisons among query and support features benefit from both rich semantics and precise localization. We demonstrate our approach primarily on few-shot video object segmentation and an adapted version on the fully supervised counterpart. In all cases, our approach outperforms the baseline and yields state-of-the-art performance. Our code is publicly available at https://github.com/MSiam/MMC-MultiscaleMemory.
Authors: Pietro Bonazzi, Thomas Ruegg, Sizhen Bian, Yawei Li, Michele Magno
Intelligent edge vision tasks face the challenge of power and latency efficiency as the computation load is normally heavy for edge platforms. This work leverages one of the first "AI in sensor" vision platforms, IMX500 by Sony, to achieve ultra-fast and ultra-low-power end-to-end edge vision applications. We evaluate the IMX500 and compare it to other edge platforms, such as the Google Coral Dev Micro and Sony Spresense, by exploring gaze estimation as a case study. We propose TinyTracker, a highly efficient, fully quantized model for 2D gaze estimation designed to maximize the performance of the edge vision systems considered in this study. TinyTracker achieves a 41x size reduction (600Kb) compared to iTracker [1] without significant loss in gaze estimation accuracy (maximum of 0.16 cm when fully quantized). TinyTracker's deployment on the Sony IMX500 vision sensor results in end-to-end latency of around 19ms. The camera takes around 17.9ms to read, process and transmit the pixels to the accelerator. The inference time of the network is 0.86ms with an additional 0.24 ms for retrieving the results from the sensor. and the overall energy consumption of the end-to-end system is 4.9 mJ, including 0.06 mJ for inference. The end-to-end study shows that IMX500 is 1.7x faster than CoralMicro (19ms vs 34.4ms) and 7x more power efficient (4.9mJ VS 34.2mJ)
Authors: Chunming He, Kai Li, Guoxia Xu, Jiangpeng Yan, Longxiang Tang, Yulun Zhang, Xiu Li, Yaowei Wang
Unpaired Medical Image Enhancement (UMIE) aims to transform a low-quality (LQ) medical image into a high-quality (HQ) one without relying on paired images for training. While most existing approaches are based on Pix2Pix/CycleGAN and are effective to some extent, they fail to explicitly use HQ information to guide the enhancement process, which can lead to undesired artifacts and structural distortions. In this paper, we propose a novel UMIE approach that avoids the above limitation of existing methods by directly encoding HQ cues into the LQ enhancement process in a variational fashion and thus model the UMIE task under the joint distribution between the LQ and HQ domains. Specifically, we extract features from an HQ image and explicitly insert the features, which are expected to encode HQ cues, into the enhancement network to guide the LQ enhancement with the variational normalization module. We train the enhancement network adversarially with a discriminator to ensure the generated HQ image falls into the HQ domain. We further propose a content-aware loss to guide the enhancement process with wavelet-based pixel-level and multi-encoder-based feature-level constraints. Additionally, as a key motivation for performing image enhancement is to make the enhanced images serve better for downstream tasks, we propose a bi-level learning scheme to optimize the UMIE task and downstream tasks cooperatively, helping generate HQ images both visually appealing and favorable for downstream tasks. Experiments on three medical datasets, including two newly collected datasets, verify that the proposed method outperforms existing techniques in terms of both enhancement quality and downstream task performance. We will make the code and the newly collected datasets publicly available for community study.
Authors: Zhaoyuan He, Yifan Yang, Shuozhe Li, Diyuan Dai, Lili Qiu
Cloud gaming is a multi-billion dollar industry. A client in cloud gaming sends its movement to the game server on the Internet, which renders and transmits the resulting video back. In order to provide a good gaming experience, a latency below 80 ms is required. This means that video rendering, encoding, transmission, decoding, and display have to finish within that time frame, which is especially challenging to achieve due to server overload, network congestion, and losses. In this paper, we propose a new method for recovering lost or corrupted video frames in cloud gaming. Unlike traditional video frame recovery, our approach uses game states to significantly enhance recovery accuracy and utilizes partially decoded frames to recover lost portions. We develop a holistic system that consists of (i) efficiently extracting game states, (ii) modifying H.264 video decoder to generate a mask to indicate which portions of video frames need recovery, and (iii) designing a novel neural network to recover either complete or partial video frames. Our approach is extensively evaluated using iPhone 12 and laptop implementations, and we demonstrate the utility of game states in the game video recovery and the effectiveness of our overall design.
Authors: Xingxing Wei, Yao Huang, Yitong Sun, Jie Yu
Recently, physical adversarial attacks have been presented to evade DNNs-based object detectors. To ensure the security, many scenarios are simultaneously deployed with visible sensors and infrared sensors, leading to the failures of these single-modal physical attacks. To show the potential risks under such scenes, we propose a unified adversarial patch to perform cross-modal physical attacks, i.e., fooling visible and infrared object detectors at the same time via a single patch. Considering different imaging mechanisms of visible and infrared sensors, our work focuses on modeling the shapes of adversarial patches, which can be captured in different modalities when they change. To this end, we design a novel boundary-limited shape optimization to achieve the compact and smooth shapes, and thus they can be easily implemented in the physical world. In addition, to balance the fooling degree between visible detector and infrared detector during the optimization process, we propose a score-aware iterative evaluation, which can guide the adversarial patch to iteratively reduce the predicted scores of the multi-modal sensors. We finally test our method against the one-stage detector: YOLOv3 and the two-stage detector: Faster RCNN. Results show that our unified patch achieves an Attack Success Rate (ASR) of 73.33% and 69.17%, respectively. More importantly, we verify the effective attacks in the physical world when visible and infrared sensors shoot the objects under various settings like different angles, distances, postures, and scenes.
Authors: Yechao Zhang, Shengshan Hu, Leo Yu Zhang, Junyu Shi, Minghui Li, Xiaogeng Liu, Wei Wan, Hai Jin
Adversarial examples (AEs) for DNNs have been shown to be transferable: AEs that successfully fool white-box surrogate models can also deceive other black-box models with different architectures. Although a bunch of empirical studies have provided guidance on generating highly transferable AEs, many of these findings lack explanations and even lead to inconsistent advice. In this paper, we take a further step towards understanding adversarial transferability, with a particular focus on surrogate aspects. Starting from the intriguing little robustness phenomenon, where models adversarially trained with mildly perturbed adversarial samples can serve as better surrogates, we attribute it to a trade-off between two predominant factors: model smoothness and gradient similarity. Our investigations focus on their joint effects, rather than their separate correlations with transferability. Through a series of theoretical and empirical analyses, we conjecture that the data distribution shift in adversarial training explains the degradation of gradient similarity. Building on these insights, we explore the impacts of data augmentation and gradient regularization on transferability and identify that the trade-off generally exists in the various training mechanisms, thus building a comprehensive blueprint for the regulation mechanism behind transferability. Finally, we provide a general route for constructing better surrogates to boost transferability which optimizes both model smoothness and gradient similarity simultaneously, e.g., the combination of input gradient regularization and sharpness-aware minimization (SAM), validated by extensive experiments. In summary, we call for attention to the united impacts of these two factors for launching effective transfer attacks, rather than optimizing one while ignoring the other, and emphasize the crucial role of manipulating surrogate models.
Authors: Sina Gholamian, Ali Vahdat
While analyzing scanned documents, handwritten text can overlay printed text. This causes difficulties during the optical character recognition (OCR) and digitization process of documents, and subsequently, hurts downstream NLP tasks. Prior research either focuses only on the binary classification of handwritten text, or performs a three-class segmentation of the document, i.e., recognition of handwritten, printed, and background pixels. This results in the assignment of the handwritten and printed overlapping pixels to only one of the classes, and thus, they are not accounted for in the other class. Thus, in this research, we develop novel approaches for addressing the challenges of handwritten and printed text segmentation with the goal of recovering text in different classes in whole, especially improving the segmentation performance on the overlapping parts. As such, to facilitate with this task, we introduce a new dataset, SignaTR6K, collected from real legal documents, as well as a new model architecture for handwritten and printed text segmentation task. Our best configuration outperforms the prior work on two different datasets by 17.9% and 7.3% on IoU scores.
Authors: Weiying Zhao, Charles-Alban Deledalle, Loïc Denis, Henri Maître, Jean-Marie Nicolas, Florence Tupin
Understanding the state of changed areas requires that precise information be given about the changes. Thus, detecting different kinds of changes is important for land surface monitoring. SAR sensors are ideal to fulfil this task, because of their all-time and all-weather capabilities, with good accuracy of the acquisition geometry and without effects of atmospheric constituents for amplitude data. In this study, we propose a simplified generalized likelihood ratio ($S_{GLR}$) method assuming that corresponding temporal pixels have the same equivalent number of looks (ENL). Thanks to the denoised data provided by a ratio-based multitemporal SAR image denoising method (RABASAR), we successfully applied this similarity test approach to compute the change areas. A new change magnitude index method and an improved spectral clustering-based change classification method are also developed. In addition, we apply the simplified generalized likelihood ratio to detect the maximum change magnitude time, and the change starting and ending times. Then, we propose to use an adaptation of the REACTIV method to visualize the detection results vividly. The effectiveness of the proposed methods is demonstrated through the processing of simulated and SAR images, and the comparison with classical techniques. In particular, numerical experiments proved that the developed method has good performances in detecting farmland area changes, building area changes, harbour area changes and flooding area changes.
Authors: Assef Ghamisi, Todd Charter, Li Ji, Maxime Rivard, Gil Lund, Homayoun Najjaran
Current defect detection systems for Automated Fibre Placement (AFP) are mostly based on end-to-end supervised learning methods requiring abundant labelled defective samples, which are not easily generated in sufficient numbers. To address this data scarcity problem, we introduce an autoencoder-based approach compatible with small datasets. Fortunately, the problem from a foundational point of view can be simplified as a binary classification between normal and abnormal samples. The proposed approach uses a depth map of the fibre layup surface, split into small windows aligned to each composite strip (tow). A subset of these windows that do not contain anomalies is passed to an autoencoder to reconstruct the input. Because the autoencoder is trained with normal samples, it produces more accurate reconstructions for these samples than for abnormal ones. Therefore, the value of reconstruction error is used as a quantitative metric for whether there are potential anomalies. These values are combined to produce an anomaly map, which can localize the manufacturing defects in the depth map. The results show that although the autoencoder is trained with a very limited number of scans, the proposed approach can produce sufficient binary classification accuracy and specify the location of the defects.
Authors: Kaveh Safavigerdini, Koundinya Nouduri, Ramakrishna Surya, Andrew Reinhard, Zach Quinlan, Filiz Bunyak, Matthew R. Maschmann, Kannappan Palaniappan
We present a pipeline for predicting mechanical properties of vertically-oriented carbon nanotube (CNT) forest images using a deep learning model for artificial intelligence (AI)-based materials discovery. Our approach incorporates an innovative data augmentation technique that involves the use of multi-layer synthetic (MLS) or quasi-2.5D images which are generated by blending 2D synthetic images. The MLS images more closely resemble 3D synthetic and real scanning electron microscopy (SEM) images of CNTs but without the computational cost of performing expensive 3D simulations or experiments. Mechanical properties such as stiffness and buckling load for the MLS images are estimated using a physics-based model. The proposed deep learning architecture, CNTNeXt, builds upon our previous CNTNet neural network, using a ResNeXt feature representation followed by random forest regression estimator. Our machine learning approach for predicting CNT physical properties by utilizing a blended set of synthetic images is expected to outperform single synthetic image-based learning when it comes to predicting mechanical properties of real scanning electron microscopy images. This has the potential to accelerate understanding and control of CNT forest self-assembly for diverse applications.
Authors: Mingyuan Fan, Cen Chen, Chengyu Wang, Wenmeng Zhou, Jun Huang
Split learning enables collaborative deep learning model training while preserving data privacy and model security by avoiding direct sharing of raw data and model details (i.e., sever and clients only hold partial sub-networks and exchange intermediate computations). However, existing research has mainly focused on examining its reliability for privacy protection, with little investigation into model security. Specifically, by exploring full models, attackers can launch adversarial attacks, and split learning can mitigate this severe threat by only disclosing part of models to untrusted servers.This paper aims to evaluate the robustness of split learning against adversarial attacks, particularly in the most challenging setting where untrusted servers only have access to the intermediate layers of the model.Existing adversarial attacks mostly focus on the centralized setting instead of the collaborative setting, thus, to better evaluate the robustness of split learning, we develop a tailored attack called SPADV, which comprises two stages: 1) shadow model training that addresses the issue of lacking part of the model and 2) local adversarial attack that produces adversarial examples to evaluate.The first stage only requires a few unlabeled non-IID data, and, in the second stage, SPADV perturbs the intermediate output of natural samples to craft the adversarial ones. The overall cost of the proposed attack process is relatively low, yet the empirical attack effectiveness is significantly high, demonstrating the surprising vulnerability of split learning to adversarial attacks.
Authors: Xiaohang Ren, Xingyu Chen, Pengfei Yao, Heung-Yeung Shum, Baoyuan Wang
The SOTA face swap models still suffer the problem of either target identity (i.e., shape) being leaked or the target non-identity attributes (i.e., background, hair) failing to be fully preserved in the final results. We show that this insufficient disentanglement is caused by two flawed designs that were commonly adopted in prior models: (1) counting on only one compressed encoder to represent both the semantic-level non-identity facial attributes(i.e., pose) and the pixel-level non-facial region details, which is contradictory to satisfy at the same time; (2) highly relying on long skip-connections between the encoder and the final generator, leaking a certain amount of target face identity into the result. To fix them, we introduce a new face swap framework called 'WSC-swap' that gets rid of skip connections and uses two target encoders to respectively capture the pixel-level non-facial region attributes and the semantic non-identity attributes in the face region. To further reinforce the disentanglement learning for the target encoder, we employ both identity removal loss via adversarial training (i.e., GAN) and the non-identity preservation loss via prior 3DMM models like [11]. Extensive experiments on both FaceForensics++ and CelebA-HQ show that our results significantly outperform previous works on a rich set of metrics, including one novel metric for measuring identity consistency that was completely neglected before.
Authors: Haofu Liao, Aruni RoyChowdhury, Weijian Li, Ankan Bansal, Yuting Zhang, Zhuowen Tu, Ravi Kumar Satzoda, R. Manmatha, Vijay Mahadevan
We present a new formulation for structured information extraction (SIE) from visually rich documents. It aims to address the limitations of existing IOB tagging or graph-based formulations, which are either overly reliant on the correct ordering of input text or struggle with decoding a complex graph. Instead, motivated by anchor-based object detectors in vision, we represent an entity as an anchor word and a bounding box, and represent entity linking as the association between anchor words. This is more robust to text ordering, and maintains a compact graph for entity linking. The formulation motivates us to introduce 1) a DOCument TRansformer (DocTr) that aims at detecting and associating entity bounding boxes in visually rich documents, and 2) a simple pre-training strategy that helps learn entity detection in the context of language. Evaluations on three SIE benchmarks show the effectiveness of the proposed formulation, and the overall approach outperforms existing solutions.
Authors: Yin Tang, Tao Chen, Xiruo Jiang, Yazhou Yao, Guo-Sen Xie, Heng-Tao Shen
Few-shot video object segmentation (FSVOS) aims to segment dynamic objects of unseen classes by resorting to a small set of support images that contain pixel-level object annotations. Existing methods have demonstrated that the domain agent-based attention mechanism is effective in FSVOS by learning the correlation between support images and query frames. However, the agent frame contains redundant pixel information and background noise, resulting in inferior segmentation performance. Moreover, existing methods tend to ignore inter-frame correlations in query videos. To alleviate the above dilemma, we propose a holistic prototype attention network (HPAN) for advancing FSVOS. Specifically, HPAN introduces a prototype graph attention module (PGAM) and a bidirectional prototype attention module (BPAM), transferring informative knowledge from seen to unseen classes. PGAM generates local prototypes from all foreground features and then utilizes their internal correlations to enhance the representation of the holistic prototypes. BPAM exploits the holistic information from support images and video frames by fusing co-attention and self-attention to achieve support-query semantic consistency and inner-frame temporal consistency. Extensive experiments on YouTube-FSVOS have been provided to demonstrate the effectiveness and superiority of our proposed HPAN method.
Authors: Siwei Yang, Hanrong Ye, Dan Xu
This paper targets the problem of multi-task dense prediction which aims to achieve simultaneous learning and inference on a bunch of multiple dense prediction tasks in a single framework. A core objective in design is how to effectively model cross-task interactions to achieve a comprehensive improvement on different tasks based on their inherent complementarity and consistency. Existing works typically design extra expensive distillation modules to perform explicit interaction computations among different task-specific features in both training and inference, bringing difficulty in adaptation for different task sets, and reducing efficiency due to clearly increased size of multi-task models. In contrast, we introduce feature-wise contrastive consistency into modeling the cross-task interactions for multi-task dense prediction. We propose a novel multi-task contrastive regularization method based on the consistency to effectively boost the representation learning of the different sub-tasks, which can also be easily generalized to different multi-task dense prediction frameworks, and costs no additional computation in the inference. Extensive experiments on two challenging datasets (i.e. NYUD-v2 and Pascal-Context) clearly demonstrate the superiority of the proposed multi-task contrastive learning approach for dense predictions, establishing new state-of-the-art performances.
Authors: Jinlong Li, Runsheng Xu, Xinyu Liu, Baolu Li, Qin Zou, Jiaqi Ma, Hongkai Yu
Due to the lack of real multi-agent data and time-consuming of labeling, existing multi-agent cooperative perception algorithms usually select the simulated sensor data for training and validating. However, the perception performance is degraded when these simulation-trained models are deployed to the real world, due to the significant domain gap between the simulated and real data. In this paper, we propose the first Simulation-to-Reality transfer learning framework for multi-agent cooperative perception using a novel Vision Transformer, named as S2R-ViT, which considers both the Implementation Gap and Feature Gap between simulated and real data. We investigate the effects of these two types of domain gaps and propose a novel uncertainty-aware vision transformer to effectively relief the Implementation Gap and an agent-based feature adaptation module with inter-agent and ego-agent discriminators to reduce the Feature Gap. Our intensive experiments on the public multi-agent cooperative perception datasets OPV2V and V2V4Real demonstrate that the proposed S2R-ViT can effectively bridge the gap from simulation to reality and outperform other methods significantly for point cloud-based 3D object detection.
Authors: Haotian Dong (1), Enhui Ma (1), Lubo Wang (1), Miaohui Wang (2), Wuyuan Xie (2), Qing Guo (3), Ping Li (4), Lingyu Liang (5), Kairui Yang (6), Di Lin (1) ((1) Tianjin University, (2) Shenzhen University, (3) A*STAR, (4) The Hong Kong Polytechnic University, (5) South China University of Technology, (6) Alibaba Damo Academy)
Semantic scene completion (SSC) requires an accurate understanding of the geometric and semantic relationships between the objects in the 3D scene for reasoning the occluded objects. The popular SSC methods voxelize the 3D objects, allowing the deep 3D convolutional network (3D CNN) to learn the object relationships from the complex scenes. However, the current networks lack the controllable kernels to model the object relationship across multiple views, where appropriate views provide the relevant information for suggesting the existence of the occluded objects. In this paper, we propose Cross-View Synthesis Transformer (CVSformer), which consists of Multi-View Feature Synthesis and Cross-View Transformer for learning cross-view object relationships. In the multi-view feature synthesis, we use a set of 3D convolutional kernels rotated differently to compute the multi-view features for each voxel. In the cross-view transformer, we employ the cross-view fusion to comprehensively learn the cross-view relationships, which form useful information for enhancing the features of individual views. We use the enhanced features to predict the geometric occupancies and semantic labels of all voxels. We evaluate CVSformer on public datasets, where CVSformer yields state-of-the-art results.
Authors: Yadan Luo, Zhuoxiao Chen, Zhen Fang, Zheng Zhang, Zi Huang, Mahsa Baktashmotlagh
Achieving a reliable LiDAR-based object detector in autonomous driving is paramount, but its success hinges on obtaining large amounts of precise 3D annotations. Active learning (AL) seeks to mitigate the annotation burden through algorithms that use fewer labels and can attain performance comparable to fully supervised learning. Although AL has shown promise, current approaches prioritize the selection of unlabeled point clouds with high uncertainty and/or diversity, leading to the selection of more instances for labeling and reduced computational efficiency. In this paper, we resort to a novel kernel coding rate maximization (KECOR) strategy which aims to identify the most informative point clouds to acquire labels through the lens of information theory. Greedy search is applied to seek desired point clouds that can maximize the minimal number of bits required to encode the latent features. To determine the uniqueness and informativeness of the selected samples from the model perspective, we construct a proxy network of the 3D detector head and compute the outer product of Jacobians from all proxy layers to form the empirical neural tangent kernel (NTK) matrix. To accommodate both one-stage (i.e., SECOND) and two-stage detectors (i.e., PVRCNN), we further incorporate the classification entropy maximization and well trade-off between detection performance and the total number of bounding boxes selected for annotation. Extensive experiments conducted on two 3D benchmarks and a 2D detection dataset evidence the superiority and versatility of the proposed approach. Our results show that approximately 44% box-level annotation costs and 26% computational time are reduced compared to the state-of-the-art AL method, without compromising detection performance.
Authors: Zhuoxiao Chen, Yadan Luo, Zi Huang, Zheng Wang, Mahsa Baktashmotlagh
Unsupervised domain adaptation (DA) with the aid of pseudo labeling techniques has emerged as a crucial approach for domain-adaptive 3D object detection. While effective, existing DA methods suffer from a substantial drop in performance when applied to a multi-class training setting, due to the co-existence of low-quality pseudo labels and class imbalance issues. In this paper, we address this challenge by proposing a novel ReDB framework tailored for learning to detect all classes at once. Our approach produces Reliable, Diverse, and class-Balanced pseudo 3D boxes to iteratively guide the self-training on a distributionally different target domain. To alleviate disruptions caused by the environmental discrepancy (e.g., beam numbers), the proposed cross-domain examination (CDE) assesses the correctness of pseudo labels by copy-pasting target instances into a source environment and measuring the prediction consistency. To reduce computational overhead and mitigate the object shift (e.g., scales and point densities), we design an overlapped boxes counting (OBC) metric that allows to uniformly downsample pseudo-labeled objects across different geometric characteristics. To confront the issue of inter-class imbalance, we progressively augment the target point clouds with a class-balanced set of pseudo-labeled target instances and source objects, which boosts recognition accuracies on both frequently appearing and rare classes. Experimental results on three benchmark datasets using both voxel-based (i.e., SECOND) and point-based 3D detectors (i.e., PointRCNN) demonstrate that our proposed ReDB approach outperforms existing 3D domain adaptation methods by a large margin, improving 23.15% mAP on the nuScenes $\rightarrow$ KITTI task.
Authors: Wuyuan Xie, Miaohui Wang, Di Lin, Boxin Shi, Jianmin Jiang
With the rapid development of high-resolution 3D vision applications, the traditional way of manipulating surface detail requires considerable memory and computing time. To address these problems, we introduce an efficient surface detail processing framework in 2D normal domain, which extracts new normal feature representations as the carrier of micro geometry structures that are illustrated both theoretically and empirically in this article. Compared with the existing state of the arts, we verify and demonstrate that the proposed normal-based representation has three important properties, including detail separability, detail transferability and detail idempotence. Finally, three new schemes are further designed for geometric surface detail processing applications, including geometric texture synthesis, geometry detail transfer, and 3D surface super-resolution. Theoretical analysis and experimental results on the latest benchmark dataset verify the effectiveness and versatility of our normal-based representation, which accepts 30 times of the input surface vertices but at the same time only takes 6.5% memory cost and 14.0% running time in comparison with existing competing algorithms.
Authors: Shuhan Tan, Boris Ivanovic, Xinshuo Weng, Marco Pavone, Philipp Kraehenbuehl
Simulation forms the backbone of modern self-driving development. Simulators help develop, test, and improve driving systems without putting humans, vehicles, or their environment at risk. However, simulators face a major challenge: They rely on realistic, scalable, yet interesting content. While recent advances in rendering and scene reconstruction make great strides in creating static scene assets, modeling their layout, dynamics, and behaviors remains challenging. In this work, we turn to language as a source of supervision for dynamic traffic scene generation. Our model, LCTGen, combines a large language model with a transformer-based decoder architecture that selects likely map locations from a dataset of maps, and produces an initial traffic distribution, as well as the dynamics of each vehicle. LCTGen outperforms prior work in both unconditional and conditional traffic scene generation in terms of realism and fidelity. Code and video will be available at https://ariostgx.github.io/lctgen.
Authors: Sahil Tyagi, Martin Swany
In distributed training, deep neural networks (DNNs) are launched over multiple workers concurrently and aggregate their local updates on each step in bulk-synchronous parallel (BSP) training. However, BSP does not linearly scale-out due to high communication cost of aggregation. To mitigate this overhead, alternatives like Federated Averaging (FedAvg) and Stale-Synchronous Parallel (SSP) either reduce synchronization frequency or eliminate it altogether, usually at the cost of lower final accuracy. In this paper, we present \texttt{SelSync}, a practical, low-overhead method for DNN training that dynamically chooses to incur or avoid communication at each step either by calling the aggregation op or applying local updates based on their significance. We propose various optimizations as part of \texttt{SelSync} to improve convergence in the context of \textit{semi-synchronous} training. Our system converges to the same or better accuracy than BSP while reducing training time by up to 14$\times$.
Authors: Lei Ma, Peng Xue, Yuning Gu, Yue Zhao, Min Zhu, Zhongxiang Ding, Dinggang Shen
In recent years, the demand for dental implants has surged, driven by their high success rates and esthetic advantages. However, accurate prediction of missing teeth for precise digital implant planning remains a challenge due to the intricate nature of dental structures and the variability in tooth loss patterns. This study presents a novel framework for accurate prediction of missing teeth in different patterns, facilitating digital implant planning. The proposed framework begins by estimating point-to-point correspondence among a dataset of dental mesh models reconstructed from CBCT images of healthy subjects. Subsequently, tooth dictionaries are constructed for each tooth type, encoding their position and shape information based on the established point-to-point correspondence. To predict missing teeth in a given dental mesh model, sparse coefficients are learned by sparsely representing adjacent teeth of the missing teeth using the corresponding tooth dictionaries. These coefficients are then applied to the dictionaries of the missing teeth to generate accurate predictions of their positions and shapes. The evaluation results on real subjects shows that our proposed framework achieves an average prediction error of 1.04mm for predictions of single missing tooth and an average prediction error of 1.33mm for the prediction of 14 missing teeth, which demonstrates its capability of accurately predicting missing teeth in various patterns. By accurately predicting missing teeth, dental professionals can improve the planning and placement of dental implants, leading to better esthetic and functional outcomes for patients undergoing dental implant procedures.
Authors: Jingyuan Yang, Qirui Huang, Tingting Ding, Dani Lischinski, Daniel Cohen-Or, Hui Huang
Visual Emotion Analysis (VEA) aims at predicting people's emotional responses to visual stimuli. This is a promising, yet challenging, task in affective computing, which has drawn increasing attention in recent years. Most of the existing work in this area focuses on feature design, while little attention has been paid to dataset construction. In this work, we introduce EmoSet, the first large-scale visual emotion dataset annotated with rich attributes, which is superior to existing datasets in four aspects: scale, annotation richness, diversity, and data balance. EmoSet comprises 3.3 million images in total, with 118,102 of these images carefully labeled by human annotators, making it five times larger than the largest existing dataset. EmoSet includes images from social networks, as well as artistic images, and it is well balanced between different emotion categories. Motivated by psychological studies, in addition to emotion category, each image is also annotated with a set of describable emotion attributes: brightness, colorfulness, scene type, object class, facial expression, and human action, which can help understand visual emotions in a precise and interpretable way. The relevance of these emotion attributes is validated by analyzing the correlations between them and visual emotion, as well as by designing an attribute module to help visual emotion recognition. We believe EmoSet will bring some key insights and encourage further research in visual emotion analysis and understanding. The data and code will be released after the publication of this work.
Authors: Dongyu Yao, Boheng Li, Run Wang, Lina Wang
To circumvent the costly pixel-wise annotations of real-world images in the semantic segmentation task, the Unsupervised Domain Adaptation (UDA) is explored to firstly train a model with the labeled source data (synthetic images) and then adapt it to the unlabeled target data (real images). Among all the techniques being studied, the self-training approach recently secures its position in domain adaptive semantic segmentation, where a model is trained with target domain pseudo-labels. Current advances have mitigated noisy pseudo-labels resulting from the domain gap. However, they still struggle with erroneous pseudo-labels near the decision boundaries of the semantic classifier. In this paper, we tackle this issue by proposing a dual-level interaction for domain adaptation (DIDA) in semantic segmentation. Explicitly, we encourage the different augmented views of the same pixel to have not only similar class prediction (semantic-level) but also akin similarity relationship respected to other pixels (instance-level). As it is impossible to keep features of all pixel instances for a dataset, we novelly design and maintain a labeled instance bank with dynamic updating strategies to selectively store the informative features of instances. Further, DIDA performs cross-level interaction with scattering and gathering techniques to regenerate more reliable pseudolabels. Our method outperforms the state-of-the-art by a notable margin, especially on confusing and long-tailed classes. Code is available at https://github.com/RainJamesY/DIDA.
Authors: Maosu Li, Yijie Wu, Anthony G.O. Yeh, Fan Xue
Many existing 3D semantic segmentation methods, deep learning in computer vision notably, claimed to achieve desired results on urban point clouds, in which the city objects are too many and diverse for people to judge qualitatively. Thus, it is significant to assess these methods quantitatively in diversified real-world urban scenes, encompassing high-rise, low-rise, high-density, and low-density urban areas. However, existing public benchmark datasets primarily represent low-rise scenes from European cities and cannot assess the methods comprehensively. This paper presents a benchmark dataset of high-rise urban point clouds, namely High-Rise, High-Density urban scenes of Hong Kong (HRHD-HK), which has been vacant for a long time. HRHD-HK arranged in 150 tiles contains 273 million colorful photogrammetric 3D points from diverse urban settings. The semantic labels of HRHD-HK include building, vegetation, road, waterbody, facility, terrain, and vehicle. To the best of our knowledge, HRHD-HK is the first photogrammetric dataset that focuses on HRHD urban areas. This paper also comprehensively evaluates eight popular semantic segmentation methods on the HRHD-HK dataset. Experimental results confirmed plenty of room for enhancing the current 3D semantic segmentation of point clouds, especially for city objects with small volumes. Our dataset is publicly available at: https://github.com/LuZaiJiaoXiaL/HRHD-HK.
Authors: Runwei Ding, Yuhang Wen, Jinfu Liu, Nan Dai, Fanyang Meng, Mengyuan Liu
Human skeletons and RGB sequences are both widely-adopted input modalities for human action recognition. However, skeletons lack appearance features and color data suffer large amount of irrelevant depiction. To address this, we introduce human parsing feature map as a novel modality, since it can selectively retain spatiotemporal features of the body parts, while filtering out noises regarding outfits, backgrounds, etc. We propose an Integrating Human Parsing and Pose Network (IPP-Net) for action recognition, which is the first to leverage both skeletons and human parsing feature maps in dual-branch approach. The human pose branch feeds compact skeletal representations of different modalities in graph convolutional network to model pose features. In human parsing branch, multi-frame body-part parsing features are extracted with human detector and parser, which is later learnt using a convolutional backbone. A late ensemble of two branches is adopted to get final predictions, considering both robust keypoints and rich semantic body-part features. Extensive experiments on NTU RGB+D and NTU RGB+D 120 benchmarks consistently verify the effectiveness of the proposed IPP-Net, which outperforms the existing action recognition methods. Our code is publicly available at https://github.com/liujf69/IPP-Net-Parsing .
Authors: Krishna Teja Chitty-Venkata, Sparsh Mittal, Murali Emani, Venkatram Vishwanath, Arun K. Somani
Recent years have seen a phenomenal rise in performance and applications of transformer neural networks. The family of transformer networks, including Bidirectional Encoder Representations from Transformer (BERT), Generative Pretrained Transformer (GPT) and Vision Transformer (ViT), have shown their effectiveness across Natural Language Processing (NLP) and Computer Vision (CV) domains. Transformer-based networks such as ChatGPT have impacted the lives of common men. However, the quest for high predictive performance has led to an exponential increase in transformers' memory and compute footprint. Researchers have proposed techniques to optimize transformer inference at all levels of abstraction. This paper presents a comprehensive survey of techniques for optimizing the inference phase of transformer networks. We survey techniques such as knowledge distillation, pruning, quantization, neural architecture search and lightweight network design at the algorithmic level. We further review hardware-level optimization techniques and the design of novel hardware accelerators for transformers. We summarize the quantitative results on the number of parameters/FLOPs and accuracy of several models/techniques to showcase the tradeoff exercised by them. We also outline future directions in this rapidly evolving field of research. We believe that this survey will educate both novice and seasoned researchers and also spark a plethora of research efforts in this field.
Authors: Tomáš Chobola, Gesine Müller, Veit Dausmann, Anton Theileis, Jan Taucher, Jan Huisken, Tingying Peng
The process of acquiring microscopic images in life sciences often results in image degradation and corruption, characterised by the presence of noise and blur, which poses significant challenges in accurately analysing and interpreting the obtained data. This paper proposes LUCYD, a novel method for the restoration of volumetric microscopy images that combines the Richardson-Lucy deconvolution formula and the fusion of deep features obtained by a fully convolutional network. By integrating the image formation process into a feature-driven restoration model, the proposed approach aims to enhance the quality of the restored images whilst reducing computational costs and maintaining a high degree of interpretability. Our results demonstrate that LUCYD outperforms the state-of-the-art methods in both synthetic and real microscopy images, achieving superior performance in terms of image quality and generalisability. We show that the model can handle various microscopy modalities and different imaging conditions by evaluating it on two different microscopy datasets, including volumetric widefield and light-sheet microscopy. Our experiments indicate that LUCYD can significantly improve resolution, contrast, and overall quality of microscopy images. Therefore, it can be a valuable tool for microscopy image restoration and can facilitate further research in various microscopy applications. We made the source code for the model accessible under https://github.com/ctom2/lucyd-deconvolution.
Authors: Mahbub Ul Alam, Jaakko Hollmén, Jón Rúnar Baldvinsson, Rahim Rahmani
The interpretability of deep neural networks has become a subject of great interest within the medical and healthcare domain. This attention stems from concerns regarding transparency, legal and ethical considerations, and the medical significance of predictions generated by these deep neural networks in clinical decision support systems. To address this matter, our study delves into the application of four well-established interpretability methods: Local Interpretable Model-agnostic Explanations (LIME), Shapley Additive exPlanations (SHAP), Gradient-weighted Class Activation Mapping (Grad-CAM), and Layer-wise Relevance Propagation (LRP). Leveraging the approach of transfer learning with a multi-label-multi-class chest radiography dataset, we aim to interpret predictions pertaining to specific pathology classes. Our analysis encompasses both single-label and multi-label predictions, providing a comprehensive and unbiased assessment through quantitative and qualitative investigations, which are compared against human expert annotation. Notably, Grad-CAM demonstrates the most favorable performance in quantitative evaluation, while the LIME heatmap segmentation visualization exhibits the highest level of medical significance. Our research highlights the strengths and limitations of these interpretability methods and suggests that a multimodal-based approach, incorporating diverse sources of information beyond chest radiography images, could offer additional insights for enhancing interpretability in the medical domain.
Authors: Yue Song, Jichao Zhang, Nicu Sebe, Wei Wang
Generative Adversarial Networks (GANs), especially the recent style-based generators (StyleGANs), have versatile semantics in the structured latent space. Latent semantics discovery methods emerge to move around the latent code such that only one factor varies during the traversal. Recently, an unsupervised method proposed a promising direction to directly use the eigenvectors of the projection matrix that maps latent codes to features as the interpretable directions. However, one overlooked fact is that the projection matrix is non-orthogonal and the number of eigenvectors is too large. The non-orthogonality would entangle semantic attributes in the top few eigenvectors, and the large dimensionality might result in meaningless variations among the directions even if the matrix is orthogonal. To avoid these issues, we propose Householder Projector, a flexible and general low-rank orthogonal matrix representation based on Householder transformations, to parameterize the projection matrix. The orthogonality guarantees that the eigenvectors correspond to disentangled interpretable semantics, while the low-rank property encourages that each identified direction has meaningful variations. We integrate our projector into pre-trained StyleGAN2/StyleGAN3 and evaluate the models on several benchmarks. Within only $1\%$ of the original training steps for fine-tuning, our projector helps StyleGANs to discover more disentangled and precise semantic attributes without sacrificing image fidelity.
Authors: Haobo Song, Soumajit Majumder, Tao Lin
Implicit models such as Deep Equilibrium Models (DEQs) have garnered significant attention in the community for their ability to train infinite layer models with elegant solution-finding procedures and constant memory footprint. However, despite several attempts, these methods are heavily constrained by model inefficiency and optimization instability. Furthermore, fair benchmarking across relevant methods for vision tasks is missing. In this work, we revisit the line of implicit models and trace them back to the original weight-tied models. Surprisingly, we observe that weight-tied models are more effective, stable, as well as efficient on vision tasks, compared to the DEQ variants. Through the lens of these simple-yet-clean weight-tied models, we further study the fundamental limits in the model capacity of such models and propose the use of distinct sparse masks to improve the model capacity. Finally, for practitioners, we offer design guidelines regarding the depth, width, and sparsity selection for weight-tied models, and demonstrate the generalizability of our insights to other learning paradigms.
Authors: Yujiao Shi, Fei Wu, Ankit Vora, Akhil Perincherry, Hongdong Li
Image retrieval-based cross-view localization methods often lead to very coarse camera pose estimation, due to the limited sampling density of the database satellite images. In this paper, we propose a method to increase the accuracy of a ground camera's location and orientation by estimating the relative rotation and translation between the ground-level image and its matched/retrieved satellite image. Our approach designs a geometry-guided cross-view transformer that combines the benefits of conventional geometry and learnable cross-view transformers to map the ground-view observations to an overhead view. Given the synthesized overhead view and observed satellite feature maps, we construct a neural pose optimizer with strong global information embedding ability to estimate the relative rotation between them. After aligning their rotations, we develop an uncertainty-guided spatial correlation to generate a probability map of the vehicle locations, from which the relative translation can be determined. Experimental results demonstrate that our method significantly outperforms the state-of-the-art. Notably, the likelihood of restricting the vehicle lateral pose to be within 1m of its Ground Truth (GT) value on the cross-view KITTI dataset has been improved from $35.54\%$ to $76.44\%$, and the likelihood of restricting the vehicle orientation to be within $1^{\circ}$ of its GT value has been improved from $19.64\%$ to $99.10\%$.
Authors: Ruipu Luo, Jiwen Zhang, Zhongyu Wei
Vision language decision making (VLDM) is a challenging multimodal task. The agent have to understand complex human instructions and complete compositional tasks involving environment navigation and object manipulation. However, the long action sequences involved in VLDM make the task difficult to learn. From an environment perspective, we find that task episodes can be divided into fine-grained \textit{units}, each containing a navigation phase and an interaction phase. Since the environment within a unit stays unchanged, we propose a novel hybrid-training framework that enables active exploration in the environment and reduces the exposure bias. Such framework leverages the unit-grained configurations and is model-agnostic. Specifically, we design a Unit-Transformer (UT) with an intrinsic recurrent state that maintains a unit-scale cross-modal memory. Through extensive experiments on the TEACH benchmark, we demonstrate that our proposed framework outperforms existing state-of-the-art methods in terms of all evaluation metrics. Overall, our work introduces a novel approach to tackling the VLDM task by breaking it down into smaller, manageable units and utilizing a hybrid-training framework. By doing so, we provide a more flexible and effective solution for multimodal decision making.
Authors: Harvey Mannering
This work presents a novel strategy to measure bias in text-to-image models. Using paired prompts that specify gender and vaguely reference an object (e.g. "a man/woman holding an item") we can examine whether certain objects are associated with a certain gender. In analysing results from Stable Diffusion, we observed that male prompts generated objects such as ties, knives, trucks, baseball bats, and bicycles more frequently. On the other hand, female prompts were more likely to generate objects such as handbags, umbrellas, bowls, bottles, and cups. We hope that the method outlined here will be a useful tool for examining bias in text-to-image models.
Authors: Sadra Safadoust, Fatma Güney
Recent work in unsupervised multi-object segmentation shows impressive results by predicting motion from a single image despite the inherent ambiguity in predicting motion without the next image. On the other hand, the set of possible motions for an image can be constrained to a low-dimensional space by considering the scene structure and moving objects in it. We propose to model pixel-wise geometry and object motion to remove ambiguity in reconstructing flow from a single image. Specifically, we divide the image into coherently moving regions and use depth to construct flow bases that best explain the observed flow in each region. We achieve state-of-the-art results in unsupervised multi-object segmentation on synthetic and real-world datasets by modeling the scene structure and object motion. Our evaluation of the predicted depth maps shows reliable performance in monocular depth estimation.
Authors: Yuying Ge, Yixiao Ge, Ziyun Zeng, Xintao Wang, Ying Shan
We present SEED, an elaborate image tokenizer that empowers Large Language Models (LLMs) with the emergent ability to SEE and Draw at the same time. Research on image tokenizers has previously reached an impasse, as frameworks employing quantized visual tokens have lost prominence due to subpar performance and convergence in multimodal comprehension (compared to BLIP-2, etc.) or generation (compared to Stable Diffusion, etc.). Despite the limitations, we remain confident in its natural capacity to unify visual and textual representations, facilitating scalable multimodal training with LLM's original recipe. In this study, we identify two crucial principles for the architecture and training of SEED that effectively ease subsequent alignment with LLMs. (1) Image tokens should be independent of 2D physical patch positions and instead be produced with a 1D causal dependency, exhibiting intrinsic interdependence that aligns with the left-to-right autoregressive prediction mechanism in LLMs. (2) Image tokens should capture high-level semantics consistent with the degree of semantic abstraction in words, and be optimized for both discriminativeness and reconstruction during the tokenizer training phase. As a result, the off-the-shelf LLM is able to perform both image-to-text and text-to-image generation by incorporating our SEED through efficient LoRA tuning. Comprehensive multimodal pretraining and instruction tuning, which may yield improved results, are reserved for future investigation. This version of SEED was trained in 5.7 days using only 64 V100 GPUs and 5M publicly available image-text pairs. Our preliminary study emphasizes the great potential of discrete visual tokens in versatile multimodal LLMs and the importance of proper image tokenizers in broader research.
Authors: Tejashwini P S, Thriveni J, Venugopal K R
Brain tumor is deliberated as one of the severe health complications which lead to decrease in life expectancy of the individuals and is also considered as a prominent cause of mortality worldwide. Therefore, timely detection and prediction of brain tumors can be helpful to prevent death rates due to brain tumors. Biomedical image analysis is a widely known solution to diagnose brain tumor. Although MRI is the current standard method for imaging tumors, its clinical usefulness is constrained by the requirement of manual segmentation which is time-consuming. Deep learning-based approaches have emerged as a promising solution to develop automated biomedical image exploration tools and the UNet architecture is commonly used for segmentation. However, the traditional UNet has limitations in terms of complexity, training, accuracy, and contextual information processing. As a result, the modified UNet architecture, which incorporates residual dense blocks, layered attention, and channel attention modules, in addition to stacked convolution, can effectively capture both coarse and fine feature information. The proposed SLCA UNet approach achieves good performance on the freely accessible Brain Tumor Segmentation (BraTS) dataset, with an average performance of 0.845, 0.845, 0.999, and 8.1 in terms of Dice, Sensitivity, Specificity, and Hausdorff95 for BraTS 2020 dataset, respectively.
Authors: Zhenqi He, Mathias Unberath, Jing Ke, Yiqing Shen
Nuclei appear small in size, yet, in real clinical practice, the global spatial information and correlation of the color or brightness contrast between nuclei and background, have been considered a crucial component for accurate nuclei segmentation. However, the field of automatic nuclei segmentation is dominated by Convolutional Neural Networks (CNNs), meanwhile, the potential of the recently prevalent Transformers has not been fully explored, which is powerful in capturing local-global correlations. To this end, we make the first attempt at a pure Transformer framework for nuclei segmentation, called TransNuSeg. Different from prior work, we decouple the challenging nuclei segmentation task into an intrinsic multi-task learning task, where a tri-decoder structure is employed for nuclei instance, nuclei edge, and clustered edge segmentation respectively. To eliminate the divergent predictions from different branches in previous work, a novel self distillation loss is introduced to explicitly impose consistency regulation between branches. Moreover, to formulate the high correlation between branches and also reduce the number of parameters, an efficient attention sharing scheme is proposed by partially sharing the self-attention heads amongst the tri-decoders. Finally, a token MLP bottleneck replaces the over-parameterized Transformer bottleneck for a further reduction in model complexity. Experiments on two datasets of different modalities, including MoNuSeg have shown that our methods can outperform state-of-the-art counterparts such as CA2.5-Net by 2-3% Dice with 30% fewer parameters. In conclusion, TransNuSeg confirms the strength of Transformer in the context of nuclei segmentation, which thus can serve as an efficient solution for real clinical practice. Code is available at https://github.com/zhenqi-he/transnuseg.
Authors: Haonan Yin, Guanlong Jiao, Qianhui Wu, Borje F. Karlsson, Biqing Huang, Chin Yew Lin
In the context of flexible manufacturing systems that are required to produce different types and quantities of products with minimal reconfiguration, this paper addresses the problem of unsupervised multi-class anomaly detection: develop a unified model to detect anomalies from objects belonging to multiple classes when only normal data is accessible. We first explore the generative-based approach and investigate latent diffusion models for reconstruction to mitigate the notorious ``identity shortcut'' issue in auto-encoder based methods. We then introduce a feature editing strategy that modifies the input feature space of the diffusion model to further alleviate ``identity shortcuts'' and meanwhile improve the reconstruction quality of normal regions, leading to fewer false positive predictions. Moreover, we are the first who pose the problem of hyperparameter selection in unsupervised anomaly detection, and propose a solution of synthesizing anomaly data for a pseudo validation set to address this problem. Extensive experiments on benchmark datasets MVTec-AD and MPDD show that the proposed LafitE, \ie, Latent Diffusion Model with Feature Editing, outperforms state-of-art methods by a significant margin in terms of average AUROC. The hyperparamters selected via our pseudo validation set are well-matched to the real test set.
Authors: Mohanad Odema, Halima Bouzidi, Hamza Ouarnoughi, Smail Niar, Mohammad Abdullah Al Faruque
Graph Neural Networks (GNNs) are becoming increasingly popular for vision-based applications due to their intrinsic capacity in modeling structural and contextual relations between various parts of an image frame. On another front, the rising popularity of deep vision-based applications at the edge has been facilitated by the recent advancements in heterogeneous multi-processor Systems on Chips (MPSoCs) that enable inference under real-time, stringent execution requirements. By extension, GNNs employed for vision-based applications must adhere to the same execution requirements. Yet contrary to typical deep neural networks, the irregular flow of graph learning operations poses a challenge to running GNNs on such heterogeneous MPSoC platforms. In this paper, we propose a novel unified design-mapping approach for efficient processing of vision GNN workloads on heterogeneous MPSoC platforms. Particularly, we develop MaGNAS, a mapping-aware Graph Neural Architecture Search framework. MaGNAS proposes a GNN architectural design space coupled with prospective mapping options on a heterogeneous SoC to identify model architectures that maximize on-device resource efficiency. To achieve this, MaGNAS employs a two-tier evolutionary search to identify optimal GNNs and mapping pairings that yield the best performance trade-offs. Through designing a supernet derived from the recent Vision GNN (ViG) architecture, we conducted experiments on four (04) state-of-the-art vision datasets using both (i) a real hardware SoC platform (NVIDIA Xavier AGX) and (ii) a performance/cost model simulator for DNN accelerators. Our experimental results demonstrate that MaGNAS is able to provide 1.57x latency speedup and is 3.38x more energy-efficient for several vision datasets executed on the Xavier MPSoC vs. the GPU-only deployment while sustaining an average 0.11% accuracy reduction from the baseline.
Authors: Hao Wu, Yulong Liu, Wenbin Gao, Xiangrong Xu
Surface defect inspection based on machine vision is often affected by uneven illumination. In order to improve the inspection rate of surface defects inspection under uneven illumination condition, this paper proposes a method for detecting surface image defects based on convolutional neural network, which is based on the adjustment of convolutional neural networks, training parameters, changing the structure of the network, to achieve the purpose of accurately identifying various defects. Experimental on defect inspection of copper strip and steel images shows that the convolutional neural network can automatically learn features without preprocessing the image, and correct identification of various types of image defects affected by uneven illumination, thus overcoming the drawbacks of traditional machine vision inspection methods under uneven illumination.
Authors: Priyansh Saxena, Akshat Maheshwari, Saumil Maheshwari
Image processing concepts can visualize the different anatomy structure of the human body. Recent advancements in the field of deep learning have made it possible to detect the growth of cancerous tissue just by a patient's brain Magnetic Resonance Imaging (MRI) scans. These methods require very high accuracy and meager false negative rates to be of any practical use. This paper presents a Convolutional Neural Network (CNN) based transfer learning approach to classify the brain MRI scans into two classes using three pre-trained models. The performances of these models are compared with each other. Experimental results show that the Resnet-50 model achieves the highest accuracy and least false negative rates as 95% and zero respectively. It is followed by VGG-16 and Inception-V3 model with an accuracy of 90% and 55% respectively.
Authors: Vadim Sushko, Dan Zhang, Juergen Gall, Anna Khoreva
Given a large dataset for training, generative adversarial networks (GANs) can achieve remarkable performance for the image synthesis task. However, training GANs in extremely low data regimes remains a challenge, as overfitting often occurs, leading to memorization or training divergence. In this work, we introduce SIV-GAN, an unconditional generative model that can generate new scene compositions from a single training image or a single video clip. We propose a two-branch discriminator architecture, with content and layout branches designed to judge internal content and scene layout realism separately from each other. This discriminator design enables synthesis of visually plausible, novel compositions of a scene, with varying content and layout, while preserving the context of the original sample. Compared to previous single image GANs, our model generates more diverse, higher quality images, while not being restricted to a single image setting. We further introduce a new challenging task of learning from a few frames of a single video. In this training setup the training images are highly similar to each other, which makes it difficult for prior GAN models to achieve a synthesis of both high quality and diversity.
Authors: Dan Liu, Xi Chen, Jie Fu, Chen Ma, Xue Liu
Inference time, model size, and accuracy are three key factors in deep model compression. Most of the existing work addresses these three key factors separately as it is difficult to optimize them all at the same time. For example, low-bit quantization aims at obtaining a faster model; weight sharing quantization aims at improving compression ratio and accuracy; and mixed-precision quantization aims at balancing accuracy and inference time. To simultaneously optimize bit-width, model size, and accuracy, we propose pruning ternary quantization (PTQ): a simple, effective, symmetric ternary quantization method. We integrate L2 normalization, pruning, and the weight decay term to reduce the weight discrepancy in the gradient estimator during quantization, thus producing highly compressed ternary weights. Our method brings the highest test accuracy and the highest compression ratio. For example, it produces a 939kb (49$\times$) 2bit ternary ResNet-18 model with only 4\% accuracy drop on the ImageNet dataset. It compresses 170MB Mask R-CNN to 5MB (34$\times$) with only 2.8\% average precision drop. Our method is verified on image classification, object detection/segmentation tasks with different network structures such as ResNet-18, ResNet-50, and MobileNetV2.
Authors: Sungmin Woo, Dogyoon Lee, Sangwon Hwang, Woojin Kim, Sangyoun Lee
Despite the remarkable success of deep learning, an optimal convolution operation on point clouds remains elusive owing to their irregular data structure. Existing methods mainly focus on designing an effective continuous kernel function that can handle an arbitrary point in continuous space. Various approaches exhibiting high performance have been proposed, but we observe that the standard pointwise feature is represented by 1D channels and can become more informative when its representation involves additional spatial feature dimensions. In this paper, we present Multidimensional Kernel Convolution (MKConv), a novel convolution operator that learns to transform the point feature representation from a vector to a multidimensional matrix. Unlike standard point convolution, MKConv proceeds via two steps. (i) It first activates the spatial dimensions of local feature representation by exploiting multidimensional kernel weights. These spatially expanded features can represent their embedded information through spatial correlation as well as channel correlation in feature space, carrying more detailed local structure information. (ii) Then, discrete convolutions are applied to the multidimensional features which can be regarded as a grid-structured matrix. In this way, we can utilize the discrete convolutions for point cloud data without voxelization that suffers from information loss. Furthermore, we propose a spatial attention module, Multidimensional Local Attention (MLA), to provide comprehensive structure awareness within the local point set by reweighting the spatial feature dimensions. We demonstrate that MKConv has excellent applicability to point cloud processing tasks including object classification, object part segmentation, and scene semantic segmentation with superior results.
Authors: Junjie Li, Yixin Zhang, Zilei Wang, Keyu Tu, Saihui Hou
Contrastive learning can largely enhance the feature discriminability in a self-supervised manner and has achieved remarkable success for various visual tasks. However, it is undesirably observed that the standard contrastive paradigm (features+$\ell_{2}$ normalization) only brings little help for domain adaptation. In this work, we delve into this phenomenon and find that the main reason is due to the class weights (weights of the final fully connected layer) which are vital for the recognition yet ignored in the optimization. To tackle this issue, we propose a simple yet powerful Probabilistic Contrastive Learning (PCL), which does not only assist in extracting discriminative features but also enforces them to be clustered around the class weights. Specifically, we break the standard contrastive paradigm by removing $\ell_{2}$ normalization and replacing the features with probabilities. In this way, PCL can enforce the probability to approximate the one-hot form, thereby reducing the deviation between the features and class weights. Benefiting from the conciseness, PCL can be well generalized to different settings. In this work, we conduct extensive experiments on five tasks and observe consistent performance gains, i.e., Unsupervised Domain Adaptation (UDA), Semi-Supervised Domain Adaptation (SSDA), Semi-Supervised Learning (SSL), UDA Detection, and UDA Semantic Segmentation. Notably, for UDA Semantic Segmentation on SYNTHIA, PCL surpasses the sophisticated CPSL-D by $>\!2\%$ in terms of mean IoU with a much smaller training cost (PCL: 1*3090, 5 days v.s. CPSL-D: 4*V100, 11 days). Code is available at https://github.com/ljjcoder/Probabilistic-Contrastive-Learning.
Authors: Gang Wu, Junjun Jiang, Xianming Liu
Contrastive learning has achieved remarkable success on various high-level tasks, but there are fewer contrastive learning-based methods proposed for low-level tasks. It is challenging to adopt vanilla contrastive learning technologies proposed for high-level visual tasks to low-level image restoration problems straightly. Because the acquired high-level global visual representations are insufficient for low-level tasks requiring rich texture and context information. In this paper, we investigate the contrastive learning-based single image super-resolution from two perspectives: positive and negative sample construction and feature embedding. The existing methods take naive sample construction approaches (e.g., considering the low-quality input as a negative sample and the ground truth as a positive sample) and adopt a prior model (e.g., pre-trained VGG model) to obtain the feature embedding. To this end, we propose a practical contrastive learning framework for SISR, named PCL-SR. We involve the generation of many informative positive and hard negative samples in frequency space. Instead of utilizing an additional pre-trained network, we design a simple but effective embedding network inherited from the discriminator network which is more task-friendly. Compared with existing benchmark methods, we re-train them by our proposed PCL-SR framework and achieve superior performance. Extensive experiments have been conducted to show the effectiveness and technical contributions of our proposed PCL-SR thorough ablation studies. The code and pre-trained models can be found at https://github.com/Aitical/PCL-SISR.
Authors: Yuting Xiao, Jiale Xu, Shenghua Gao
Benefiting from the continuous representation ability, deep implicit functions can represent a shape at infinite resolution. However, extracting high-resolution iso-surface from an implicit function requires forward-propagating a network with a large number of parameters for numerous query points, thus preventing the generation speed. Inspired by the Taylor series, we propose Taylo3DNet to accelerate the inference of implicit shape representations. Taylor3DNet exploits a set of discrete landmark points and their corresponding Taylor series coefficients to represent the implicit field of a 3D shape, and the number of landmark points is independent of the resolution of the iso-surface extraction. Once the coefficients corresponding to the landmark points are predicted, the network evaluation for each query point can be simplified as a low-order Taylor series calculation with several nearest landmark points. Based on this efficient representation, our Taylor3DNet achieves a significantly faster inference speed than classical network-based implicit functions. We evaluate our approach on reconstruction tasks with various input types, and the results demonstrate that our approach can improve the inference speed by a large margin without sacrificing the performance compared with state-of-the-art baselines.
Authors: Mennatullah Siam, Konstantinos G. Derpanis, Richard P. Wildes
Few-shot video object segmentation (FS-VOS) aims at segmenting video frames using a few labelled examples of classes not seen during initial training. In this paper, we present a simple but effective temporal transductive inference (TTI) approach that leverages temporal consistency in the unlabelled video frames during few-shot inference. Key to our approach is the use of both global and local temporal constraints. The objective of the global constraint is to learn consistent linear classifiers for novel classes across the image sequence, whereas the local constraint enforces the proportion of foreground/background regions in each frame to be coherent across a local temporal window. These constraints act as spatiotemporal regularizers during the transductive inference to increase temporal coherence and reduce overfitting on the few-shot support set. Empirically, our model outperforms state-of-the-art meta-learning approaches in terms of mean intersection over union on YouTube-VIS by 2.8%. In addition, we introduce improved benchmarks that are exhaustively labelled (i.e. all object occurrences are labelled, unlike the currently available), and present a more realistic evaluation paradigm that targets data distribution shift between training and testing sets. Our empirical results and in-depth analysis confirm the added benefits of the proposed spatiotemporal regularizers to improve temporal coherence and overcome certain overfitting scenarios.
Authors: Xuehai He, Chunyuan Li, Pengchuan Zhang, Jianwei Yang, Xin Eric Wang
In computer vision, it has achieved great transfer learning performance via adapting large-scale pretrained vision models (e.g., vision transformers) to downstream tasks. Common approaches for model adaptation either update all model parameters or leverage linear probes. In this paper, we aim to study parameter-efficient model adaptation strategies for vision transformers on the image classification task. We formulate efficient model adaptation as a subspace training problem and perform a comprehensive benchmarking over different efficient adaptation methods. We conduct an empirical study on each efficient model adaptation method focusing on its performance alongside parameter cost. Furthermore, we propose a parameter-efficient model adaptation framework, which first selects submodules by measuring local intrinsic dimensions and then projects them into subspace for further decomposition via a novel Kronecker Adaptation (KAdaptation) method. We analyze and compare our method with a diverse set of baseline model adaptation methods (including state-of-the-art methods for pretrained language models). Our method performs the best in terms of the tradeoff between accuracy and parameter efficiency across 20 image classification datasets under the few-shot setting and 7 image classification datasets under the full-shot setting.
Authors: Cameron Smith, Hong-Xing Yu, Sergey Zakharov, Fredo Durand, Joshua B. Tenenbaum, Jiajun Wu, Vincent Sitzmann
Neural scene representations, both continuous and discrete, have recently emerged as a powerful new paradigm for 3D scene understanding. Recent efforts have tackled unsupervised discovery of object-centric neural scene representations. However, the high cost of ray-marching, exacerbated by the fact that each object representation has to be ray-marched separately, leads to insufficiently sampled radiance fields and thus, noisy renderings, poor framerates, and high memory and time complexity during training and rendering. Here, we propose to represent objects in an object-centric, compositional scene representation as light fields. We propose a novel light field compositor module that enables reconstructing the global light field from a set of object-centric light fields. Dubbed Compositional Object Light Fields (COLF), our method enables unsupervised learning of object-centric neural scene representations, state-of-the-art reconstruction and novel view synthesis performance on standard datasets, and rendering and training speeds at orders of magnitude faster than existing 3D approaches.
Authors: Yefei He, Luoming Zhang, Weijia Wu, Hong Zhou
Binary neural networks leverage $\mathrm{Sign}$ function to binarize weights and activations, which require gradient estimators to overcome its non-differentiability and will inevitably bring gradient errors during backpropagation. Although many hand-designed soft functions have been proposed as gradient estimators to better approximate gradients, their mechanism is not clear and there are still huge performance gaps between binary models and their full-precision counterparts. To address these issues and reduce gradient error, we propose to tackle network binarization as a binary classification problem and use a multi-layer perceptron (MLP) as the classifier in the forward pass and gradient estimator in the backward pass. Benefiting from the MLP's theoretical capability to fit any continuous function, it can be adaptively learned to binarize networks and backpropagate gradients without any prior knowledge of soft functions. From this perspective, we further empirically justify that even a simple linear function can outperform previous complex soft functions. Extensive experiments demonstrate that the proposed method yields surprising performance both in image classification and human pose estimation tasks. Specifically, we achieve $65.7\%$ top-1 accuracy of ResNet-34 on ImageNet dataset, with an absolute improvement of $2.6\%$. Moreover, we take binarization as a lightweighting approach for pose estimation models and propose well-designed binary pose estimation networks SBPN and BHRNet. When evaluating on the challenging Microsoft COCO keypoint dataset, the proposed method enables binary networks to achieve a mAP of up to $60.6$ for the first time. Experiments conducted on real platforms demonstrate that BNN achieves a better balance between performance and computational complexity, especially when computational resources are extremely low.
Authors: Asha Rani, Pankaj Yadav, Yashaswi Verma
Autism, also known as Autism Spectrum Disorder (or ASD), is a neurological disorder. Its main symptoms include difficulty in (verbal and/or non-verbal) communication, and rigid/repetitive behavior. These symptoms are often indistinguishable from a normal (control) individual, due to which this disorder remains undiagnosed in early childhood leading to delayed treatment. Since the learning curve is steep during the initial age, an early diagnosis of autism could allow to take adequate interventions at the right time, which might positively affect the growth of an autistic child. Further, the traditional methods of autism diagnosis require multiple visits to a specialized psychiatrist, however this process can be time-consuming. In this paper, we present a learning based approach to automate autism diagnosis using simple and small action video clips of subjects. This task is particularly challenging because the amount of annotated data available is small, and the variations among samples from the two categories (ASD and control) are generally indistinguishable. This is also evident from poor performance of a binary classifier learned using the cross-entropy loss on top of a baseline encoder. To address this, we adopt contrastive feature learning in both self supervised and supervised learning frameworks, and show that these can lead to a significant increase in the prediction accuracy of a binary classifier on this task. We further validate this by conducting thorough experimental analyses under different set-ups on two publicly available datasets.
Authors: Jiankai Sun, Yan Xu, Mingyu Ding, Hongwei Yi, Chen Wang, Jingdong Wang, Liangjun Zhang, Mac Schwager
Neural Radiance Fields (NeRFs) have become a widely-applied scene representation technique in recent years, showing advantages for robot navigation and manipulation tasks. To further advance the utility of NeRFs for robotics, we propose a transformer-based framework, NeRF-Loc, to extract 3D bounding boxes of objects in NeRF scenes. NeRF-Loc takes a pre-trained NeRF model and camera view as input and produces labeled, oriented 3D bounding boxes of objects as output. Using current NeRF training tools, a robot can train a NeRF environment model in real-time and, using our algorithm, identify 3D bounding boxes of objects of interest within the NeRF for downstream navigation or manipulation tasks. Concretely, we design a pair of paralleled transformer encoder branches, namely the coarse stream and the fine stream, to encode both the context and details of target objects. The encoded features are then fused together with attention layers to alleviate ambiguities for accurate object localization. We have compared our method with conventional RGB(-D) based methods that take rendered RGB images and depths from NeRFs as inputs. Our method is better than the baselines.
Authors: Ruinan Jin, Xiaoxiao Li
Deep Learning-based image synthesis techniques have been applied in healthcare research for generating medical images to support open research and augment medical datasets. Training generative adversarial neural networks (GANs) usually require large amounts of training data. Federated learning (FL) provides a way of training a central model using distributed data while keeping raw data locally. However, given that the FL server cannot access the raw data, it is vulnerable to backdoor attacks, an adversarial by poisoning training data. Most backdoor attack strategies focus on classification models and centralized domains. It is still an open question if the existing backdoor attacks can affect GAN training and, if so, how to defend against the attack in the FL setting. In this work, we investigate the overlooked issue of backdoor attacks in federated GANs (FedGANs). The success of this attack is subsequently determined to be the result of some local discriminators overfitting the poisoned data and corrupting the local GAN equilibrium, which then further contaminates other clients when averaging the generator's parameters and yields high generator loss. Therefore, we proposed FedDetect, an efficient and effective way of defending against the backdoor attack in the FL setting, which allows the server to detect the client's adversarial behavior based on their losses and block the malicious clients. Our extensive experiments on two medical datasets with different modalities demonstrate the backdoor attack on FedGANs can result in synthetic images with low fidelity. After detecting and suppressing the detected malicious clients using the proposed defense strategy, we show that FedGANs can synthesize high-quality medical datasets (with labels) for data augmentation to improve classification models' performance.
Authors: Muhammad Asif Khan, Hamid Menouar, Ridha Hamila
Video surveillance using drones is both convenient and efficient due to the ease of deployment and unobstructed movement of drones in many scenarios. An interesting application of drone-based video surveillance is to estimate crowd densities (both pedestrians and vehicles) in public places. Deep learning using convolution neural networks (CNNs) is employed for automatic crowd counting and density estimation using images and videos. However, the performance and accuracy of such models typically depend upon the model architecture i.e., deeper CNN models improve accuracy at the cost of increased inference time. In this paper, we propose a novel crowd density estimation model for drones (DroneNet) using Self-organized Operational Neural Networks (Self-ONN). Self-ONN provides efficient learning capabilities with lower computational complexity as compared to CNN-based models. We tested our algorithm on two drone-view public datasets. Our evaluation shows that the proposed DroneNet shows superior performance on an equivalent CNN-based model.
Authors: Bowen Li, Ziyuan Huang, Junjie Ye, Yiming Li, Sebastian Scherer, Hang Zhao, Changhong Fu
Visual object tracking is essential to intelligent robots. Most existing approaches have ignored the online latency that can cause severe performance degradation during real-world processing. Especially for unmanned aerial vehicles (UAVs), where robust tracking is more challenging and onboard computation is limited, the latency issue can be fatal. In this work, we present a simple framework for end-to-end latency-aware tracking, i.e., end-to-end predictive visual tracking (PVT++). Unlike existing solutions that naively append Kalman Filters after trackers, PVT++ can be jointly optimized, so that it takes not only motion information but can also leverage the rich visual knowledge in most pre-trained tracker models for robust prediction. Besides, to bridge the training-evaluation domain gap, we propose a relative motion factor, empowering PVT++ to generalize to the challenging and complex UAV tracking scenes. These careful designs have made the small-capacity lightweight PVT++ a widely effective solution. Additionally, this work presents an extended latency-aware evaluation benchmark for assessing an any-speed tracker in the online setting. Empirical results on a robotic platform from the aerial perspective show that PVT++ can achieve significant performance gain on various trackers and exhibit higher accuracy than prior solutions, largely mitigating the degradation brought by latency.
Authors: Guanlin Li, Guowen Xu, Tianwei Zhang
In this paper, we consider the instance segmentation task on a long-tailed dataset, which contains label noise, i.e., some of the annotations are incorrect. There are two main reasons making this case realistic. First, datasets collected from real world usually obey a long-tailed distribution. Second, for instance segmentation datasets, as there are many instances in one image and some of them are tiny, it is easier to introduce noise into the annotations. Specifically, we propose a new dataset, which is a large vocabulary long-tailed dataset containing label noise for instance segmentation. Furthermore, we evaluate previous proposed instance segmentation algorithms on this dataset. The results indicate that the noise in the training dataset will hamper the model in learning rare categories and decrease the overall performance, and inspire us to explore more effective approaches to address this practical challenge. The code and dataset are available in https://github.com/GuanlinLee/Noisy-LVIS.
Authors: Sriram Balasubramanian, Soheil Feizi
The ability to remove features from the input of machine learning models is very important to understand and interpret model predictions. However, this is non-trivial for vision models since masking out parts of the input image typically causes large distribution shifts. This is because the baseline color used for masking (typically grey or black) is out of distribution. Furthermore, the shape of the mask itself can contain unwanted signals which can be used by the model for its predictions. Recently, there has been some progress in mitigating this issue (called missingness bias) in image masking for vision transformers. In this work, we propose a new masking method for CNNs we call layer masking in which the missingness bias caused by masking is reduced to a large extent. Intuitively, layer masking applies a mask to intermediate activation maps so that the model only processes the unmasked input. We show that our method (i) is able to eliminate or minimize the influence of the mask shape or color on the output of the model, and (ii) is much better than replacing the masked region by black or grey for input perturbation based interpretability techniques like LIME. Thus, layer masking is much less affected by missingness bias than other masking strategies. We also demonstrate how the shape of the mask may leak information about the class, thus affecting estimates of model reliance on class-relevant features derived from input masking. Furthermore, we discuss the role of data augmentation techniques for tackling this problem, and argue that they are not sufficient for preventing model reliance on mask shape. The code for this project is publicly available at https://github.com/SriramB-98/layer_masking
Authors: Vishaal Udandarao, Ankush Gupta, Samuel Albanie
Contrastive Language-Image Pre-training (CLIP) has emerged as a simple yet effective way to train large-scale vision-language models. CLIP demonstrates impressive zero-shot classification and retrieval on diverse downstream tasks. However, to leverage its full potential, fine-tuning still appears to be necessary. Fine-tuning the entire CLIP model can be resource-intensive and unstable. Moreover, recent methods that aim to circumvent this need for fine-tuning still require access to images from the target distribution. In this paper, we pursue a different approach and explore the regime of training-free "name-only transfer" in which the only knowledge we possess about the downstream task comprises the names of downstream target categories. We propose a novel method, SuS-X, consisting of two key building blocks -- SuS and TIP-X, that requires neither intensive fine-tuning nor costly labelled data. SuS-X achieves state-of-the-art zero-shot classification results on 19 benchmark datasets. We further show the utility of TIP-X in the training-free few-shot setting, where we again achieve state-of-the-art results over strong training-free baselines. Code is available at https://github.com/vishaal27/SuS-X.
Authors: Muhammad Asif Khan, Hamid Menouar, Ridha Hamila
Density estimation is one of the most widely used methods for crowd counting in which a deep learning model learns from head-annotated crowd images to estimate crowd density in unseen images. Typically, the learning performance of the model is highly impacted by the accuracy of the annotations and inaccurate annotations may lead to localization and counting errors during prediction. A significant amount of works exist on crowd counting using perfectly labelled datasets but none of these explore the impact of annotation errors on the model accuracy. In this paper, we investigate the impact of imperfect labels (both noisy and missing labels) on crowd counting accuracy. We propose a system that automatically generates imperfect labels using a deep learning model (called annotator) which are then used to train a new crowd counting model (target model). Our analysis on two crowd counting models and two benchmark datasets shows that the proposed scheme achieves accuracy closer to that of the model trained with perfect labels showing the robustness of crowd models to annotation errors.
Authors: Muhammad Asif Khan, Ridha Hamila, Hamid Menouar
Deep learning models require an enormous amount of data for training. However, recently there is a shift in machine learning from model-centric to data-centric approaches. In data-centric approaches, the focus is to refine and improve the quality of the data to improve the learning performance of the models rather than redesigning model architectures. In this paper, we propose CLIP i.e., Curriculum Learning with Iterative data Pruning. CLIP combines two data-centric approaches i.e., curriculum learning and dataset pruning to improve the model learning accuracy and convergence speed. The proposed scheme applies loss-aware dataset pruning to iteratively remove the least significant samples and progressively reduces the size of the effective dataset in the curriculum learning training. Extensive experiments performed on crowd density estimation models validate the notion behind combining the two approaches by reducing the convergence time and improving generalization. To our knowledge, the idea of data pruning as an embedded process in curriculum learning is novel.
Authors: Seong Hun Lee, Javier Civera
One of the main limitations of the commonly used Absolute Trajectory Error (ATE) is that it is highly sensitive to outliers. As a result, in the presence of just a few outliers, it often fails to reflect the varying accuracy as the inlier trajectory error or the number of outliers varies. In this work, we propose an alternative error metric for evaluating the accuracy of the reconstructed camera trajectory. Our metric, named Discernible Trajectory Error (DTE), is computed in five steps: (1) Shift the ground-truth and estimated trajectories such that both of their geometric medians are located at the origin. (2) Rotate the estimated trajectory such that it minimizes the sum of geodesic distances between the corresponding camera orientations. (3) Scale the estimated trajectory such that the median distance of the cameras to their geometric median is the same as that of the ground truth. (4) Compute and winsorize the distances between the corresponding cameras. (5) Obtain the DTE by taking the average of the mean and the root-mean-square (RMS) of the winsorized distances. This metric is an attractive alternative to the ATE, in that it is capable of discerning the varying trajectory accuracy as the inlier trajectory error or the number of outliers varies. Using the similar idea, we also propose a novel rotation error metric, named Discernible Rotation Error (DRE), which has similar advantages to the DTE. Furthermore, we propose a simple yet effective method for calibrating the camera-to-marker rotation, which is needed for the computation of our metrics. Our methods are verified through extensive simulations.
Authors: Wenye Lin, Yifeng Ding, Zhixiong Cao, Hai-tao Zheng
Recent research has reported a performance degradation in self-supervised contrastive learning for specially designed efficient networks, such as MobileNet and EfficientNet. A common practice to address this problem is to introduce a pretrained contrastive teacher model and train the lightweight networks with distillation signals generated by the teacher. However, it is time and resource consuming to pretrain a teacher model when it is not available. In this work, we aim to establish a stronger baseline for lightweight contrastive models without using a pretrained teacher model. Specifically, we show that the optimal recipe for efficient models is different from that of larger models, and using the same training settings as ResNet50, as previous research does, is inappropriate. Additionally, we observe a common issu e in contrastive learning where either the positive or negative views can be noisy, and propose a smoothed version of InfoNCE loss to alleviate this problem. As a result, we successfully improve the linear evaluation results from 36.3\% to 62.3\% for MobileNet-V3-Large and from 42.2\% to 65.8\% for EfficientNet-B0 on ImageNet, closing the accuracy gap to ResNet50 with $5\times$ fewer parameters. We hope our research will facilitate the usage of lightweight contrastive models.
Authors: Po-Yao Huang, Vasu Sharma, Hu Xu, Chaitanya Ryali, Haoqi Fan, Yanghao Li, Shang-Wen Li, Gargi Ghosh, Jitendra Malik, Christoph Feichtenhofer
We present Masked Audio-Video Learners (MAViL) to train audio-visual representations. Our approach learns with three complementary forms of self-supervision: (1) reconstruction of masked audio and video input data, (2) intra- and inter-modal contrastive learning with masking, and (3) self-training by reconstructing joint audio-video contextualized features learned from the first two objectives. Pre-training with MAViL not only enables the model to perform well in audio-visual classification and retrieval tasks but also improves representations of each modality in isolation, without using information from the other modality for fine-tuning or inference. Empirically, MAViL sets a new state-of-the-art on AudioSet (53.1 mAP) and VGGSound (67.1% accuracy). For the first time, a self-supervised audio-visual model outperforms ones that use external supervision on these benchmarks.
Authors: Zekun Qian, Ruize Han, Wei Feng, Feifan Wang, Song Wang
We tackle a new problem of multi-view camera and subject registration in the bird's eye view (BEV) without pre-given camera calibration. This is a very challenging problem since its only input is several RGB images from different first-person views (FPVs) for a multi-person scene, without the BEV image and the calibration of the FPVs, while the output is a unified plane with the localization and orientation of both the subjects and cameras in a BEV. We propose an end-to-end framework solving this problem, whose main idea can be divided into following parts: i) creating a view-transform subject detection module to transform the FPV to a virtual BEV including localization and orientation of each pedestrian, ii) deriving a geometric transformation based method to estimate camera localization and view direction, i.e., the camera registration in a unified BEV, iii) making use of spatial and appearance information to aggregate the subjects into the unified BEV. We collect a new large-scale synthetic dataset with rich annotations for evaluation. The experimental results show the remarkable effectiveness of our proposed method.
Authors: Surat Teerapittayanon, Marcus Comiter, Brad McDanel, H.T. Kung
We propose StitchNet, a novel neural network creation paradigm that stitches together fragments (one or more consecutive network layers) from multiple pre-trained neural networks. StitchNet allows the creation of high-performing neural networks without the large compute and data requirements needed under traditional model creation processes via backpropagation training. We leverage Centered Kernel Alignment (CKA) as a compatibility measure to efficiently guide the selection of these fragments in composing a network for a given task tailored to specific accuracy needs and computing resource constraints. We then show that these fragments can be stitched together to create neural networks with comparable accuracy to traditionally trained networks at a fraction of computing resource and data requirements. Finally, we explore a novel on-the-fly personalized model creation and inference application enabled by this new paradigm.
Authors: Peng Tu, Xu Xie, Guo AI, Yuexiang Li, Yawen Huang, Yefeng Zheng
Efficient detectors for edge devices are often optimized for parameters or speed count metrics, which remain in weak correlation with the energy of detectors.
However, some vision applications of convolutional neural networks, such as always-on surveillance cameras, are critical for energy constraints.
This paper aims to serve as a baseline by designing detectors to reach tradeoffs between energy and performance from two perspectives:
1) We extensively analyze various CNNs to identify low-energy architectures, including selecting activation functions, convolutions operators, and feature fusion structures on necks. These underappreciated details in past work seriously affect the energy consumption of detectors;
2) To break through the dilemmatic energy-performance problem, we propose a balanced detector driven by energy using discovered low-energy components named \textit{FemtoDet}.
In addition to the novel construction, we improve FemtoDet by considering convolutions and training strategy optimizations.
Specifically, we develop a new instance boundary enhancement (IBE) module for convolution optimization to overcome the contradiction between the limited capacity of CNNs and detection tasks in diverse spatial representations, and propose a recursive warm-restart (RecWR) for optimizing training strategy to escape the sub-optimization of light-weight detectors by considering the data shift produced in popular augmentations.
As a result, FemtoDet with only 68.77k parameters achieves a competitive score of 46.3 AP50 on PASCAL VOC and 1.11 W $\&$ 64.47 FPS on Qualcomm Snapdragon 865 CPU platforms.
Extensive experiments on COCO and TJU-DHD datasets indicate that the proposed method achieves competitive results in diverse scenes.
Authors: Conghao Xiong, Hao Chen, Joseph J.Y. Sung, Irwin King
Multiple Instance Learning (MIL) and transformers are increasingly popular in histopathology Whole Slide Image (WSI) classification. However, unlike human pathologists who selectively observe specific regions of histopathology tissues under different magnifications, most methods do not incorporate multiple resolutions of the WSIs, hierarchically and attentively, thereby leading to a loss of focus on the WSIs and information from other resolutions. To resolve this issue, we propose a Hierarchical Attention-Guided Multiple Instance Learning framework to fully exploit the WSIs. This framework can dynamically and attentively discover the discriminative regions across multiple resolutions of the WSIs. Within this framework, an Integrated Attention Transformer is proposed to further enhance the performance of the transformer and obtain a more holistic WSI (bag) representation. This transformer consists of multiple Integrated Attention Modules, which is the combination of a transformer layer and an aggregation module that produces a bag representation based on every instance representation in that bag. The experimental results show that our method achieved state-of-the-art performances on multiple datasets, including Camelyon16, TCGA-RCC, TCGA-NSCLC, and an in-house IMGC dataset. The code is available at https://github.com/BearCleverProud/HAG-MIL.
Authors: Zhijian Liu, Xinyu Yang, Haotian Tang, Shang Yang, Song Han
Transformer, as an alternative to CNN, has been proven effective in many modalities (e.g., texts and images). For 3D point cloud transformers, existing efforts focus primarily on pushing their accuracy to the state-of-the-art level. However, their latency lags behind sparse convolution-based models (3x slower), hindering their usage in resource-constrained, latency-sensitive applications (such as autonomous driving). This inefficiency comes from point clouds' sparse and irregular nature, whereas transformers are designed for dense, regular workloads. This paper presents FlatFormer to close this latency gap by trading spatial proximity for better computational regularity. We first flatten the point cloud with window-based sorting and partition points into groups of equal sizes rather than windows of equal shapes. This effectively avoids expensive structuring and padding overheads. We then apply self-attention within groups to extract local features, alternate sorting axis to gather features from different directions, and shift windows to exchange features across groups. FlatFormer delivers state-of-the-art accuracy on Waymo Open Dataset with 4.6x speedup over (transformer-based) SST and 1.4x speedup over (sparse convolutional) CenterPoint. This is the first point cloud transformer that achieves real-time performance on edge GPUs and is faster than sparse convolutional methods while achieving on-par or even superior accuracy on large-scale benchmarks.
Authors: Yongsheng Mei, Guru Venkataramani, Tian Lan
Learning with multiple modalities is crucial for automated brain tumor segmentation from magnetic resonance imaging data. Explicitly optimizing the common information shared among all modalities (e.g., by maximizing the total correlation) has been shown to achieve better feature representations and thus enhance the segmentation performance. However, existing approaches are oblivious to partial common information shared by subsets of the modalities. In this paper, we show that identifying such partial common information can significantly boost the discriminative power of image segmentation models. In particular, we introduce a novel concept of partial common information mask (PCI-mask) to provide a fine-grained characterization of what partial common information is shared by which subsets of the modalities. By solving a masked correlation maximization and simultaneously learning an optimal PCI-mask, we identify the latent microstructure of partial common information and leverage it in a self-attention module to selectively weight different feature representations in multi-modal data. We implement our proposed framework on the standard U-Net. Our experimental results on the Multi-modal Brain Tumor Segmentation Challenge (BraTS) datasets outperform those of state-of-the-art segmentation baselines, with validation Dice similarity coefficients of 0.920, 0.897, 0.837 for the whole tumor, tumor core, and enhancing tumor on BraTS-2020.
Authors: Ling-Hao Chen, Jiawei Zhang, Yewen Li, Yiren Pang, Xiaobo Xia, Tongliang Liu
Human motion prediction is a classical problem in computer vision and computer graphics, which has a wide range of practical applications. Previous effects achieve great empirical performance based on an encoding-decoding style. The methods of this style work by first encoding previous motions to latent representations and then decoding the latent representations into predicted motions. However, in practice, they are still unsatisfactory due to several issues, including complicated loss constraints, cumbersome training processes, and scarce switch of different categories of motions in prediction. In this paper, to address the above issues, we jump out of the foregoing style and propose a novel framework from a new perspective. Specifically, our framework works in a masked completion fashion. In the training stage, we learn a motion diffusion model that generates motions from random noise. In the inference stage, with a denoising procedure, we make motion prediction conditioning on observed motions to output more continuous and controllable predictions. The proposed framework enjoys promising algorithmic properties, which only needs one loss in optimization and is trained in an end-to-end manner. Additionally, it accomplishes the switch of different categories of motions effectively, which is significant in realistic tasks, e.g., the animation task. Comprehensive experiments on benchmarks confirm the superiority of the proposed framework. The project page is available at https://lhchen.top/Human-MAC.
Authors: James Urquhart Allingham, Jie Ren, Michael W Dusenberry, Xiuye Gu, Yin Cui, Dustin Tran, Jeremiah Zhe Liu, Balaji Lakshminarayanan
Contrastively trained text-image models have the remarkable ability to perform zero-shot classification, that is, classifying previously unseen images into categories that the model has never been explicitly trained to identify. However, these zero-shot classifiers need prompt engineering to achieve high accuracy. Prompt engineering typically requires hand-crafting a set of prompts for individual downstream tasks. In this work, we aim to automate this prompt engineering and improve zero-shot accuracy through prompt ensembling. In particular, we ask "Given a large pool of prompts, can we automatically score the prompts and ensemble those that are most suitable for a particular downstream dataset, without needing access to labeled validation data?". We demonstrate that this is possible. In doing so, we identify several pathologies in a naive prompt scoring method where the score can be easily overconfident due to biases in pre-training and test data, and we propose a novel prompt scoring method that corrects for the biases. Using our proposed scoring method to create a weighted average prompt ensemble, our method outperforms equal average ensemble, as well as hand-crafted prompts, on ImageNet, 4 of its variants, and 11 fine-grained classification benchmarks, all while being fully automatic, optimization-free, and not requiring access to labeled validation data.
Authors: Chinedu Innocent Nwoye, Tong Yu, Saurav Sharma, Aditya Murali, Deepak Alapatt, Armine Vardazaryan, Kun Yuan, Jonas Hajek, Wolfgang Reiter, Amine Yamlahi, Finn-Henri Smidt, Xiaoyang Zou, Guoyan Zheng, Bruno Oliveira, Helena R. Torres, Satoshi Kondo, Satoshi Kasai, Felix Holm, Ege Özsoy, Shuangchun Gui, Han Li, Sista Raviteja, Rachana Sathish, Pranav Poudel, Binod Bhattarai, Ziheng Wang, Guo Rui, Melanie Schellenberg, João L. Vilaça, Tobias Czempiel, Zhenkun Wang, Debdoot Sheet, Shrawan Kumar Thapa, Max Berniker, Patrick Godau, Pedro Morais, Sudarshan Regmi, Thuy Nuong Tran, Jaime Fonseca, Jan-Hinrich Nölke, Estevão Lima, Eduard Vazquez, Lena Maier-Hein, Nassir Navab, Pietro Mascagni, Barbara Seeliger, Cristians Gonzalez, Didier Mutter, Nicolas Padoy
Formalizing surgical activities as triplets of the used instruments, actions performed, and target anatomies is becoming a gold standard approach for surgical activity modeling. The benefit is that this formalization helps to obtain a more detailed understanding of tool-tissue interaction which can be used to develop better Artificial Intelligence assistance for image-guided surgery. Earlier efforts and the CholecTriplet challenge introduced in 2021 have put together techniques aimed at recognizing these triplets from surgical footage. Estimating also the spatial locations of the triplets would offer a more precise intraoperative context-aware decision support for computer-assisted intervention. This paper presents the CholecTriplet2022 challenge, which extends surgical action triplet modeling from recognition to detection. It includes weakly-supervised bounding box localization of every visible surgical instrument (or tool), as the key actors, and the modeling of each tool-activity in the form of <instrument, verb, target> triplet. The paper describes a baseline method and 10 new deep learning algorithms presented at the challenge to solve the task. It also provides thorough methodological comparisons of the methods, an in-depth analysis of the obtained results across multiple metrics, visual and procedural challenges; their significance, and useful insights for future research directions and applications in surgery.
Authors: Yimin Dou, Kewen Li, Wenjun Lv, Timing Li, Hongjie Duan, Zhifeng Xu
The automated interpretation and inversion of seismic data have advanced significantly with the development of Deep Learning (DL) methods. However, these methods often require numerous costly well logs, limiting their application only to mature or synthetic data. This paper presents ContrasInver, a method that achieves seismic inversion using as few as two or three well logs, significantly reducing current requirements. In ContrasInver, we propose three key innovations to address the challenges of applying semi-supervised learning to regression tasks with ultra-sparse labels. The Multi-dimensional Sample Generation (MSG) technique pioneers a paradigm for sample generation in multi-dimensional inversion. It produces a large number of diverse samples from a single well, while establishing lateral continuity in seismic data. MSG yields substantial improvements over current techniques, even without the use of semi-supervised learning. The Region-Growing Training (RGT) strategy leverages the inherent continuity of seismic data, effectively propagating accuracy from closer to more distant regions based on the proximity of well logs. The Impedance Vectorization Projection (IVP) vectorizes impedance values and performs semi-supervised learning in a compressed space. We demonstrated that the Jacobian matrix derived from this space can filter out some outlier components in pseudo-label vectors, thereby solving the value confusion issue in semi-supervised regression learning. In the experiments, ContrasInver achieved state-of-the-art performance in the synthetic data SEAM I. In the field data with two or three well logs, only the methods based on the components proposed in this paper were able to achieve reasonable results. It's the first data-driven approach yielding reliable results on the Netherlands F3 and Delft, using only three and two well logs respectively.
Authors: Yanjun Liu, Wenming Yang
Indoor 3D object detection is an essential task in single image scene understanding, impacting spatial cognition fundamentally in visual reasoning. Existing works on 3D object detection from a single image either pursue this goal through independent predictions of each object or implicitly reason over all possible objects, failing to harness relational geometric information between objects. To address this problem, we propose a dynamic sparse graph pipeline named Explicit3D based on object geometry and semantics features. Taking the efficiency into consideration, we further define a relatedness score and design a novel dynamic pruning algorithm followed by a cluster sampling method for sparse scene graph generation and updating. Furthermore, our Explicit3D introduces homogeneous matrices and defines new relative loss and corner loss to model the spatial difference between target pairs explicitly. Instead of using ground-truth labels as direct supervision, our relative and corner loss are derived from the homogeneous transformation, which renders the model to learn the geometric consistency between objects. The experimental results on the SUN RGB-D dataset demonstrate that our Explicit3D achieves better performance balance than the-state-of-the-art.
Authors: Hyunsu Kim, Gayoung Lee, Yunjey Choi, Jin-Hwa Kim, Jun-Yan Zhu
Image blending aims to combine multiple images seamlessly. It remains challenging for existing 2D-based methods, especially when input images are misaligned due to differences in 3D camera poses and object shapes. To tackle these issues, we propose a 3D-aware blending method using generative Neural Radiance Fields (NeRF), including two key components: 3D-aware alignment and 3D-aware blending. For 3D-aware alignment, we first estimate the camera pose of the reference image with respect to generative NeRFs and then perform 3D local alignment for each part. To further leverage 3D information of the generative NeRF, we propose 3D-aware blending that directly blends images on the NeRF's latent representation space, rather than raw pixel space. Collectively, our method outperforms existing 2D baselines, as validated by extensive quantitative and qualitative evaluations with FFHQ and AFHQ-Cat.
Authors: Moon Ye-Bin, Dongmin Choi, Yongjin Kwon, Junsik Kim, Tae-Hyun Oh
We address a weakly-supervised low-shot instance segmentation, an annotation-efficient training method to deal with novel classes effectively. Since it is an under-explored problem, we first investigate the difficulty of the problem and identify the performance bottleneck by conducting systematic analyses of model components and individual sub-tasks with a simple baseline model. Based on the analyses, we propose ENInst with sub-task enhancement methods: instance-wise mask refinement for enhancing pixel localization quality and novel classifier composition for improving classification accuracy. Our proposed method lifts the overall performance by enhancing the performance of each sub-task. We demonstrate that our ENInst is 7.5 times more efficient in achieving comparable performance to the existing fully-supervised few-shot models and even outperforms them at times.
Authors: Felix Friedrich, Manuel Brack, Lukas Struppek, Dominik Hintersdorf, Patrick Schramowski, Sasha Luccioni, Kristian Kersting
Generative AI models have recently achieved astonishing results in quality and are consequently employed in a fast-growing number of applications. However, since they are highly data-driven, relying on billion-sized datasets randomly scraped from the internet, they also suffer from degenerated and biased human behavior, as we demonstrate. In fact, they may even reinforce such biases. To not only uncover but also combat these undesired effects, we present a novel strategy, called Fair Diffusion, to attenuate biases after the deployment of generative text-to-image models. Specifically, we demonstrate shifting a bias, based on human instructions, in any direction yielding arbitrarily new proportions for, e.g., identity groups. As our empirical evaluation demonstrates, this introduced control enables instructing generative image models on fairness, with no data filtering and additional training required.
Authors: Jiyuan Liu, Bingyi Lu, Mingkang Xiong, Tao Zhang, Huilin Xiong
Deep neural networks (DNNs) are known to be vulnerable to adversarial examples, which are usually designed artificially to fool DNNs, but rarely exist in real-world scenarios. In this paper, we study the adversarial examples caused by raindrops, to demonstrate that there exist plenty of natural phenomena being able to work as adversarial attackers to DNNs. Moreover, we present a new approach to generate adversarial raindrops, denoted as AdvRD, using the generative adversarial network (GAN) technique to simulate natural raindrops. The images crafted by our AdvRD look very similar to the real-world raindrop images, statistically close to the distribution of true raindrop images, and more importantly, can perform strong adversarial attack to the state-of-the-art DNN models. On the other side, we show that the adversarial training using our AdvRD images can significantly improve the robustness of DNNs to the real-world raindrop attacks. Extensive experiments are carried out to demonstrate that the images crafted by AdvRD are visually and statistically close to the natural raindrop images, can work as strong attackers to DNN models, and also help improve the robustness of DNNs to raindrop attacks.
Authors: Dejun Feng, Hongyu Chen, Suning Liu, Xingyu Shen, Ziyang Liao, Yakun Xie, Jun Zhu
Salient object detection in optical remote sensing image (ORSI-SOD) has gradually attracted attention thanks to the development of deep learning (DL) and salient object detection in natural scene image (NSI-SOD). However, NSI and ORSI are different in many aspects, such as large coverage, complex background, and large differences in target types and scales. Therefore, a new dedicated method is needed for ORSI-SOD. In addition, existing methods do not pay sufficient attention to the boundary of the object, and the completeness of the final saliency map still needs improvement. To address these issues, we propose a novel method called Dual Feedback Attention Framework via Boundary-Aware Auxiliary and Progressive Semantic Optimization (DFA-BASO). First, Boundary Protection Calibration (BPC) module is proposed to reduce the loss of edge position information during forward propagation and suppress noise in low-level features. Second, a Dual Feature Feedback Complementary (DFFC) module is proposed based on BPC module. It aggregates boundary-semantic dual features and provides effective feedback to coordinate features across different layers. Finally, a Strong Semantic Feedback Refinement (SSFR) module is proposed to obtain more complete saliency maps. This module further refines feature representation and eliminates feature differences through a unique feedback mechanism. Extensive experiments on two public datasets show that DFA-BASO outperforms 15 state-of-the-art methods. Furthermore, this paper strongly demonstrates the true contribution of DFA-BASO to ORSI-SOD by in-depth analysis of the visualization figure. All codes can be found at https://github.com/YUHsss/DFA-BASO.
Authors: Hritik Bansal, Nishad Singhi, Yu Yang, Fan Yin, Aditya Grover, Kai-Wei Chang
Multimodal contrastive pretraining has been used to train multimodal representation models, such as CLIP, on large amounts of paired image-text data. However, previous studies have revealed that such models are vulnerable to backdoor attacks. Specifically, when trained on backdoored examples, CLIP learns spurious correlations between the embedded backdoor trigger and the target label, aligning their representations in the joint embedding space. Injecting even a small number of poisoned examples, such as 75 examples in 3 million pretraining data, can significantly manipulate the model's behavior, making it difficult to detect or unlearn such correlations. To address this issue, we propose CleanCLIP, a finetuning framework that weakens the learned spurious associations introduced by backdoor attacks by independently re-aligning the representations for individual modalities. We demonstrate that unsupervised finetuning using a combination of multimodal contrastive and unimodal self-supervised objectives for individual modalities can significantly reduce the impact of the backdoor attack. Additionally, we show that supervised finetuning on task-specific labeled image data removes the backdoor trigger from the CLIP vision encoder. We show empirically that CleanCLIP maintains model performance on benign examples while erasing a range of backdoor attacks on multimodal contrastive learning. The code and checkpoints are available at https://github.com/nishadsinghi/CleanCLIP.
Authors: Xinge Yang, Qiang Fu, Mohammed Elhoseiny, Wolfgang Heidrich
Computer vision methods for depth estimation usually use simple camera models with idealized optics. For modern machine learning approaches, this creates an issue when attempting to train deep networks with simulated data, especially for focus-sensitive tasks like Depth-from-Focus. In this work, we investigate the domain gap caused by off-axis aberrations that will affect the decision of the best-focused frame in a focal stack. We then explore bridging this domain gap through aberration-aware training (AAT). Our approach involves a lightweight network that models lens aberrations at different positions and focus distances, which is then integrated into the conventional network training pipeline. We evaluate the generality of pretrained models on both synthetic and real-world data. Our experimental results demonstrate that the proposed AAT scheme can improve depth estimation accuracy without fine-tuning the model or modifying the network architecture.
Authors: Takumi Harada, Hiroyuki Sakai
Prediction of image memorability has attracted interest in various fields. Consequently, the prediction accuracy of convolutional neural network (CNN) models has been approaching the empirical upper bound estimated based on human consistency. However, identifying which feature representations embedded in CNN models are responsible for the high memorability prediction accuracy remains an open question. To tackle this problem, we sought to identify memorability-related feature representations in CNN models using brain similarity. Specifically, memorability prediction accuracy and brain similarity were examined across 16,860 layers in 64 CNN models pretrained for object recognition. A clear tendency was observed in this comprehensive analysis that layers with high memorability prediction accuracy had higher brain similarity with the inferior temporal (IT) cortex, which is the highest stage in the ventral visual pathway. Furthermore, fine-tuning of the 64 CNN models for memorability prediction revealed that brain similarity with the IT cortex at the penultimate layer positively correlated with the memorability prediction accuracy of the models. This analysis also showed that the best fine-tuned model provided accuracy comparable to state-of-the-art CNN models developed for memorability prediction. Overall, the results of this study indicated that the CNN models' great success in predicting memorability relies on feature representation acquisition, similar to the IT cortex. This study advances our understanding of feature representations and their use in predicting image memorability.
Authors: Kunyang Han, Yong Liu, Jun Hao Liew, Henghui Ding, Yunchao Wei, Jiajun Liu, Yitong Wang, Yansong Tang, Yujiu Yang, Jiashi Feng, Yao Zhao
Recent advancements in pre-trained vision-language models, such as CLIP, have enabled the segmentation of arbitrary concepts solely from textual inputs, a process commonly referred to as open-vocabulary semantic segmentation (OVS). However, existing OVS techniques confront a fundamental challenge: the trained classifier tends to overfit on the base classes observed during training, resulting in suboptimal generalization performance to unseen classes. To mitigate this issue, recent studies have proposed the use of an additional frozen pre-trained CLIP for classification. Nonetheless, this approach incurs heavy computational overheads as the CLIP vision encoder must be repeatedly forward-passed for each mask, rendering it impractical for real-world applications. To address this challenge, our objective is to develop a fast OVS model that can perform comparably or better without the extra computational burden of the CLIP image encoder during inference. To this end, we propose a core idea of preserving the generalizable representation when fine-tuning on known classes. Specifically, we introduce a text diversification strategy that generates a set of synonyms for each training category, which prevents the learned representation from collapsing onto specific known category names. Additionally, we employ a text-guided knowledge distillation method to preserve the generalizable knowledge of CLIP. Extensive experiments demonstrate that our proposed model achieves robust generalization performance across various datasets. Furthermore, we perform a preliminary exploration of open-vocabulary video segmentation and present a benchmark that can facilitate future open-vocabulary research in the video domain.
Authors: Duolikun Danier, Fan Zhang, David Bull
Existing works on video frame interpolation (VFI) mostly employ deep neural networks trained to minimize the L1 or L2 distance between their outputs and ground-truth frames. Despite recent advances, existing VFI methods tend to produce perceptually inferior results, particularly for challenging scenarios including large motions and dynamic textures. Towards developing perceptually-oriented VFI methods, we propose latent diffusion model-based VFI, LDMVFI. This approaches the VFI problem from a generative perspective by formulating it as a conditional generation problem. As the first effort to address VFI using latent diffusion models, we rigorously benchmark our method following the common evaluation protocol adopted in the existing VFI literature. Our quantitative experiments and user study indicate that LDMVFI is able to interpolate video content with superior perceptual quality compared to the state of the art, even in the high-resolution regime. Our source code will be made available here.
Authors: Andrey Voynov, Qinghao Chu, Daniel Cohen-Or, Kfir Aberman
We introduce an Extended Textual Conditioning space in text-to-image models, referred to as $P+$. This space consists of multiple textual conditions, derived from per-layer prompts, each corresponding to a layer of the denoising U-net of the diffusion model.
We show that the extended space provides greater disentangling and control over image synthesis. We further introduce Extended Textual Inversion (XTI), where the images are inverted into $P+$, and represented by per-layer tokens.
We show that XTI is more expressive and precise, and converges faster than the original Textual Inversion (TI) space. The extended inversion method does not involve any noticeable trade-off between reconstruction and editability and induces more regular inversions.
We conduct a series of extensive experiments to analyze and understand the properties of the new space, and to showcase the effectiveness of our method for personalizing text-to-image models. Furthermore, we utilize the unique properties of this space to achieve previously unattainable results in object-style mixing using text-to-image models. Project page: https://prompt-plus.github.io
Authors: Pengfei Zhu, Mengshi Qi, Xia Li, Weijian Li, Huadong Ma
Predicting attention regions of interest is an important yet challenging task for self-driving systems. Existing methodologies rely on large-scale labeled traffic datasets that are labor-intensive to obtain. Besides, the huge domain gap between natural scenes and traffic scenes in current datasets also limits the potential for model training. To address these challenges, we are the first to introduce an unsupervised way to predict self-driving attention by uncertainty modeling and driving knowledge integration. Our approach's Uncertainty Mining Branch (UMB) discovers commonalities and differences from multiple generated pseudo-labels achieved from models pre-trained on natural scenes by actively measuring the uncertainty. Meanwhile, our Knowledge Embedding Block (KEB) bridges the domain gap by incorporating driving knowledge to adaptively refine the generated pseudo-labels. Quantitative and qualitative results with equivalent or even more impressive performance compared to fully-supervised state-of-the-art approaches across all three public datasets demonstrate the effectiveness of the proposed method and the potential of this direction. The code will be made publicly available.
Authors: Luis C. Garcia-Peraza-Herrera, Sebastien Ourselin, Tom Vercauteren
The performance of deep learning (DL) algorithms is heavily influenced by the quantity and the quality of the annotated data. However, in Surgical Data Science, access to it is limited. It is thus unsurprising that substantial research efforts are made to develop methods aiming at mitigating the scarcity of annotated SDS data. In parallel, an increasing number of Computer Assisted Interventions (CAI) datasets are being released, although the scale of these remain limited. On these premises, data curation is becoming a key element of many SDS research endeavors. Surgical video datasets are demanding to curate and would benefit from dedicated support tools. In this work, we propose to summarize surgical videos into storyboards or collages of representative frames to ease visualization, annotation, and processing. Video summarization is well-established for natural images. However, state-of-the-art methods typically rely on models trained on human-made annotations, few methods have been evaluated on surgical videos, and the availability of software packages for the task is limited. We present videosum, an easy-to-use and open-source Python library to generate storyboards from surgical videos that contains a variety of unsupervised methods.
Authors: Wenhang Ge, Tao Hu, Haoyu Zhao, Shu Liu, Ying-Cong Chen
Neural implicit surface learning has shown significant progress in multi-view 3D reconstruction, where an object is represented by multilayer perceptrons that provide continuous implicit surface representation and view-dependent radiance. However, current methods often fail to accurately reconstruct reflective surfaces, leading to severe ambiguity. To overcome this issue, we propose Ref-NeuS, which aims to reduce ambiguity by attenuating the effect of reflective surfaces. Specifically, we utilize an anomaly detector to estimate an explicit reflection score with the guidance of multi-view context to localize reflective surfaces. Afterward, we design a reflection-aware photometric loss that adaptively reduces ambiguity by modeling rendered color as a Gaussian distribution, with the reflection score representing the variance. We show that together with a reflection direction-dependent radiance, our model achieves high-quality surface reconstruction on reflective surfaces and outperforms the state-of-the-arts by a large margin. Besides, our model is also comparable on general surfaces.
Authors: Ameya Prabhu, Hasan Abed Al Kader Hammoud, Puneet Dokania, Philip H.S. Torr, Ser-Nam Lim, Bernard Ghanem, Adel Bibi
Continual Learning (CL) aims to sequentially train models on streams of incoming data that vary in distribution by preserving previous knowledge while adapting to new data. Current CL literature focuses on restricted access to previously seen data, while imposing no constraints on the computational budget for training. This is unreasonable for applications in-the-wild, where systems are primarily constrained by computational and time budgets, not storage. We revisit this problem with a large-scale benchmark and analyze the performance of traditional CL approaches in a compute-constrained setting, where effective memory samples used in training can be implicitly restricted as a consequence of limited computation. We conduct experiments evaluating various CL sampling strategies, distillation losses, and partial fine-tuning on two large-scale datasets, namely ImageNet2K and Continual Google Landmarks V2 in data incremental, class incremental, and time incremental settings. Through extensive experiments amounting to a total of over 1500 GPU-hours, we find that, under compute-constrained setting, traditional CL approaches, with no exception, fail to outperform a simple minimal baseline that samples uniformly from memory. Our conclusions are consistent in a different number of stream time steps, e.g., 20 to 200, and under several computational budgets. This suggests that most existing CL methods are particularly too computationally expensive for realistic budgeted deployment. Code for this project is available at: https://github.com/drimpossible/BudgetCL.
Authors: Sanket Kachole, Xiaoqian Huang, Fariborz Baghaei Naeini, Rajkumar Muthusamy, Dimitrios Makris, Yahya Zweiri
Object segmentation for robotic grasping under dynamic conditions often faces challenges such as occlusion, low light conditions, motion blur and object size variance. To address these challenges, we propose a Deep Learning network that fuses two types of visual signals, event-based data and RGB frame data. The proposed Bimodal SegNet network has two distinct encoders, one for each signal input and a spatial pyramidal pooling with atrous convolutions. Encoders capture rich contextual information by pooling the concatenated features at different resolutions while the decoder obtains sharp object boundaries. The evaluation of the proposed method undertakes five unique image degradation challenges including occlusion, blur, brightness, trajectory and scale variance on the Event-based Segmentation (ESD) Dataset. The evaluation results show a 6-10\% segmentation accuracy improvement over state-of-the-art methods in terms of mean intersection over the union and pixel accuracy. The model code is available at https://github.com/sanket0707/Bimodal-SegNet.git
Authors: Xi Chen, Shuang Li, Ser-Nam Lim, Antonio Torralba, Hengshuang Zhao
Open-vocabulary image segmentation is attracting increasing attention due to its critical applications in the real world. Traditional closed-vocabulary segmentation methods are not able to characterize novel objects, whereas several recent open-vocabulary attempts obtain unsatisfactory results, i.e., notable performance reduction on the closed vocabulary and massive demand for extra data. To this end, we propose OPSNet, an omnipotent and data-efficient framework for Open-vocabulary Panoptic Segmentation. Specifically, the exquisitely designed Embedding Modulation module, together with several meticulous components, enables adequate embedding enhancement and information exchange between the segmentation model and the visual-linguistic well-aligned CLIP encoder, resulting in superior segmentation performance under both open- and closed-vocabulary settings with much fewer need of additional data. Extensive experimental evaluations are conducted across multiple datasets (e.g., COCO, ADE20K, Cityscapes, and PascalContext) under various circumstances, where the proposed OPSNet achieves state-of-the-art results, which demonstrates the effectiveness and generality of the proposed approach. The code and trained models will be made publicly available.
Authors: Zhendong Yang, Ailing Zeng, Zhe Li, Tianke Zhang, Chun Yuan, Yu Li
Knowledge Distillation (KD) uses the teacher's prediction logits as soft labels to guide the student, while self-KD does not need a real teacher to require the soft labels. This work unifies the formulations of the two tasks by decomposing and reorganizing the generic KD loss into a Normalized KD (NKD) loss and customized soft labels for both target class (image's category) and non-target classes named Universal Self-Knowledge Distillation (USKD). We decompose the KD loss and find the non-target loss from it forces the student's non-target logits to match the teacher's, but the sum of the two non-target logits is different, preventing them from being identical. NKD normalizes the non-target logits to equalize their sum. It can be generally used for KD and self-KD to better use the soft labels for distillation loss. USKD generates customized soft labels for both target and non-target classes without a teacher. It smooths the target logit of the student as the soft target label and uses the rank of the intermediate feature to generate the soft non-target labels with Zipf's law. For KD with teachers, our NKD achieves state-of-the-art performance on CIFAR-100 and ImageNet datasets, boosting the ImageNet Top-1 accuracy of ResNet18 from 69.90% to 71.96% with a ResNet-34 teacher. For self-KD without teachers, USKD is the first self-KD method that can be effectively applied to both CNN and ViT models with negligible additional time and memory cost, resulting in new state-of-the-art results, such as 1.17% and 0.55% accuracy gains on ImageNet for MobileNet and DeiT-Tiny, respectively. Our codes are available at https://github.com/yzd-v/cls_KD.
Authors: Zhixi Cai, Shreya Ghosh, Abhinav Dhall, Tom Gedeon, Kalin Stefanov, Munawar Hayat
Most deepfake detection methods focus on detecting spatial and/or spatio-temporal changes in facial attributes and are centered around the binary classification task of detecting whether a video is real or fake. This is because available benchmark datasets contain mostly visual-only modifications present in the entirety of the video. However, a sophisticated deepfake may include small segments of audio or audio-visual manipulations that can completely change the meaning of the video content. To addresses this gap, we propose and benchmark a new dataset, Localized Audio Visual DeepFake (LAV-DF), consisting of strategic content-driven audio, visual and audio-visual manipulations. The proposed baseline method, Boundary Aware Temporal Forgery Detection (BA-TFD), is a 3D Convolutional Neural Network-based architecture which effectively captures multimodal manipulations. We further improve (i.e. BA-TFD+) the baseline method by replacing the backbone with a Multiscale Vision Transformer and guide the training process with contrastive, frame classification, boundary matching and multimodal boundary matching loss functions. The quantitative analysis demonstrates the superiority of BA-TFD+ on temporal forgery localization and deepfake detection tasks using several benchmark datasets including our newly proposed dataset. The dataset, models and code are available at https://github.com/ControlNet/LAV-DF.
Authors: Sheng Yan, Haoqiang Wang, Xin Du, Mengyuan Liu, Hong Liu
Cross-modal retrieval has become a prominent research topic in computer vision and natural language processing with advances made in image-text and video-text retrieval technologies. However, cross-modal retrieval between human motion sequences and text has not garnered sufficient attention despite the extensive application value it holds, such as aiding virtual reality applications in better understanding users' actions and language. This task presents several challenges, including joint modeling of the two modalities, demanding the understanding of person-centered information from text, and learning behavior features from 3D human motion sequences. Previous work on motion data modeling mainly relied on autoregressive feature extractors that may forget previous information, while we propose an innovative model that includes simple yet powerful transformer-based motion and text encoders, which can learn representations from the two different modalities and capture long-term dependencies. Furthermore, the overlap of the same atomic actions of different human motions can cause semantic conflicts, leading us to explore a new triplet loss function, MildTriple Loss. it leverages the similarity between samples in intra-modal space to guide soft-hard negative sample mining in the joint embedding space to train the triplet loss and reduce the violation caused by false negative samples. We evaluated our model and method on the latest HumanML3D and KIT Motion-Language datasets, achieving a 62.9\% recall for motion retrieval and a 71.5\% recall for text retrieval (based on R@10) on the HumanML3D dataset. Our code is available at https://github.com/eanson023/rehamot.
Authors: Julio Ivan Davila Carrazco, Suvarna Kishorkumar Kadam, Pietro Morerio, Alessio Del Bue, Vittorio Murino
In this paper, we introduce a novel framework for the challenging problem of One-Shot Unsupervised Domain Adaptation (OSUDA), which aims to adapt to a target domain with only a single unlabeled target sample. Unlike existing approaches that rely on large labeled source and unlabeled target data, our Target-driven One-Shot UDA (TOS-UDA) approach employs a learnable augmentation strategy guided by the target sample's style to align the source distribution with the target distribution. Our method consists of three modules: an augmentation module, a style alignment module, and a classifier. Unlike existing methods, our augmentation module allows for strong transformations of the source samples, and the style of the single target sample available is exploited to guide the augmentation by ensuring perceptual similarity. Furthermore, our approach integrates augmentation with style alignment, eliminating the need for separate pre-training on additional datasets. Our method outperforms or performs comparably to existing OS-UDA methods on the Digits and DomainNet benchmarks.
Authors: Nico Montali, John Lambert, Paul Mougin, Alex Kuefler, Nick Rhinehart, Michelle Li, Cole Gulino, Tristan Emrich, Zoey Yang, Shimon Whiteson, Brandyn White, Dragomir Anguelov
Simulation with realistic, interactive agents represents a key task for autonomous vehicle software development. In this work, we introduce the Waymo Open Sim Agents Challenge (WOSAC). WOSAC is the first public challenge to tackle this task and propose corresponding metrics. The goal of the challenge is to stimulate the design of realistic simulators that can be used to evaluate and train a behavior model for autonomous driving. We outline our evaluation methodology, present results for a number of different baseline simulation agent methods, and analyze several submissions to the 2023 competition which ran from March 16, 2023 to May 23, 2023. The WOSAC evaluation server remains open for submissions and we discuss open problems for the task.
Authors: Fangda Li, Zhiqiang Hu, Wen Chen, Avinash Kak
Hematoxylin and Eosin (H&E) staining is a widely used sample preparation procedure for enhancing the saturation of tissue sections and the contrast between nuclei and cytoplasm in histology images for medical diagnostics. However, various factors, such as the differences in the reagents used, result in high variability in the colors of the stains actually recorded. This variability poses a challenge in achieving generalization for machine-learning based computer-aided diagnostic tools. To desensitize the learned models to stain variations, we propose the Generative Stain Augmentation Network (G-SAN) -- a GAN-based framework that augments a collection of cell images with simulated yet realistic stain variations. At its core, G-SAN uses a novel and highly computationally efficient Laplacian Pyramid (LP) based generator architecture, that is capable of disentangling stain from cell morphology. Through the task of patch classification and nucleus segmentation, we show that using G-SAN-augmented training data provides on average 15.7% improvement in F1 score and 7.3% improvement in panoptic quality, respectively. Our code is available at https://github.com/lifangda01/GSAN-Demo.
Authors: Tianlun Zheng, Zhineng Chen, BingChen Huang, Wei Zhang, Yu-Gang Jiang
Traditional Multilingual Text Recognition (MLTR) usually targets a fixed set of languages and thus struggles to handle newly added languages or adapt to ever-changing class distributions. In this paper, we introduce the Incremental Multilingual Text Recognition (IMLTR) task in the incremental learning setting, where new language data comes in batches. Compared to generic incremental learning, IMLTR is even more challenging as it suffers from rehearsal-imbalance (uneven distribution of sample characters in the rehearsal set). To address this issue, we propose a Multiplexed Routing Network (MRN), where a series of recognizers is trained for each language. Subsequently, a language predictor is adopted to weigh the recognizers for voting. Since the recognizers are derived from the original model, MRN effectively reduces the reliance on older data and is better suited for rehearsal-imbalance. We extensively evaluate MRN on MLT17 and MLT19 datasets, outperforming existing state-of-the-art methods by a large margin, i.e., accuracy improvement ranging from 10.3% to 27.4% under different settings.
Authors: Ruiyang Li, Fujun Yang, Xianjie Liu, Hongwei Shi
Prosthetic Joint Infection (PJI) is a prevalent and severe complication characterized by high diagnostic challenges. Currently, a unified diagnostic standard incorporating both computed tomography (CT) images and numerical text data for PJI remains unestablished, owing to the substantial noise in CT images and the disparity in data volume between CT images and text data. This study introduces a diagnostic method, HGT, based on deep learning and multimodal techniques. It effectively merges features from CT scan images and patients' numerical text data via a Unidirectional Selective Attention (USA) mechanism and a graph convolutional network (GCN)-based feature fusion network. We evaluated the proposed method on a custom-built multimodal PJI dataset, assessing its performance through ablation experiments and interpretability evaluations. Our method achieved an accuracy (ACC) of 91.4\% and an area under the curve (AUC) of 95.9\%, outperforming recent multimodal approaches by 2.9\% in ACC and 2.2\% in AUC, with a parameter count of only 68M. Notably, the interpretability results highlighted our model's strong focus and localization capabilities at lesion sites. This proposed method could provide clinicians with additional diagnostic tools to enhance accuracy and efficiency in clinical practice.
Authors: Yuechen Zhang, Jinbo Xing, Eric Lo, Jiaya Jia
Recent diffusion model advancements have enabled high-fidelity images to be generated using text prompts. However, a domain gap exists between generated images and real-world images, which poses a challenge in generating high-quality variations of real-world images. Our investigation uncovers that this domain gap originates from a latents' distribution gap in different diffusion processes. To address this issue, we propose a novel inference pipeline called Real-world Image Variation by ALignment (RIVAL) that utilizes diffusion models to generate image variations from a single image exemplar. Our pipeline enhances the generation quality of image variations by aligning the image generation process to the source image's inversion chain. Specifically, we demonstrate that step-wise latent distribution alignment is essential for generating high-quality variations. To attain this, we design a cross-image self-attention injection for feature interaction and a step-wise distribution normalization to align the latent features. Incorporating these alignment processes into a diffusion model allows RIVAL to generate high-quality image variations without further parameter optimization. Our experimental results demonstrate that our proposed approach outperforms existing methods with respect to semantic-condition similarity and perceptual quality. Furthermore, this generalized inference pipeline can be easily applied to other diffusion-based generation tasks, such as image-conditioned text-to-image generation and example-based image inpainting.
Authors: Tejasvee Bisen, Mohammed Javed, Shashank Kirtania, P. Nagabhushan
For any digital application with document images such as retrieval, the classification of document images becomes an essential stage. Conventionally for the purpose, the full versions of the documents, that is the uncompressed document images make the input dataset, which poses a threat due to the big volume required to accommodate the full versions of the documents. Therefore, it would be novel, if the same classification task could be accomplished directly (with some partial decompression) with the compressed representation of documents in order to make the whole process computationally more efficient. In this research work, a novel deep learning model, DWT CompCNN is proposed for classification of documents that are compressed using High Throughput JPEG 2000 (HTJ2K) algorithm. The proposed DWT-CompCNN comprises of five convolutional layers with filter sizes of 16, 32, 64, 128, and 256 consecutively for each increasing layer to improve learning from the wavelet coefficients extracted from the compressed images. Experiments are performed on two benchmark datasets- Tobacco-3482 and RVL-CDIP, which demonstrate that the proposed model is time and space efficient, and also achieves a better classification accuracy in compressed domain.
Authors: Md. Simul Hasan Talukder, Rejwan Bin Sulaiman, Mohammad Raziuddin Chowdhury, Musarrat Saberin Nipun, Taminul Islam
Potatoes are the third-largest food crop globally, but their production frequently encounters difficulties because of aggressive pest infestations. The aim of this study is to investigate the various types and characteristics of these pests and propose an efficient PotatoPestNet AI-based automatic potato pest identification system. To accomplish this, we curated a reliable dataset consisting of eight types of potato pests. We leveraged the power of transfer learning by employing five customized, pre-trained transfer learning models: CMobileNetV2, CNASLargeNet, CXception, CDenseNet201, and CInceptionV3, in proposing a robust PotatoPestNet model to accurately classify potato pests. To improve the models' performance, we applied various augmentation techniques, incorporated a global average pooling layer, and implemented proper regularization methods. To further enhance the performance of the models, we utilized random search (RS) optimization for hyperparameter tuning. This optimization technique played a significant role in fine-tuning the models and achieving improved performance. We evaluated the models both visually and quantitatively, utilizing different evaluation metrics. The robustness of the models in handling imbalanced datasets was assessed using the Receiver Operating Characteristic (ROC) curve. Among the models, the Customized Tuned Inception V3 (CTInceptionV3) model, optimized through random search, demonstrated outstanding performance. It achieved the highest accuracy (91%), precision (91%), recall (91%), and F1-score (91%), showcasing its superior ability to accurately identify and classify potato pests.
Authors: Bongjin Koo, Julien Martel, Ariana Peck, Axel Levy, Frédéric Poitevin, Nina Miolane
Cryogenic electron microscopy (cryo-EM) has transformed structural biology by allowing to reconstruct 3D biomolecular structures up to near-atomic resolution. However, the 3D reconstruction process remains challenging, as the 3D structures may exhibit substantial shape variations, while the 2D image acquisition suffers from a low signal-to-noise ratio, requiring to acquire very large datasets that are time-consuming to process. Current reconstruction methods are precise but computationally expensive, or faster but lack a physically-plausible model of large molecular shape variations. To fill this gap, we propose CryoChains that encodes large deformations of biomolecules via rigid body transformation of their chains, while representing their finer shape variations with the normal mode analysis framework of biophysics. Our synthetic data experiments on the human GABA\textsubscript{B} and heat shock protein show that CryoChains gives a biophysically-grounded quantification of the heterogeneous conformations of biomolecules, while reconstructing their 3D molecular structures at an improved resolution compared to the current fastest, interpretable deep learning method.
Authors: Alex Falcon, Giuseppe Serra
In this report, we present the technical details of our submission to the EPIC-Kitchens-100 Multi-Instance Retrieval Challenge 2023. To participate in the challenge, we ensembled two models trained with two different loss functions on 25% of the training data. Our submission, visible on the public leaderboard, obtains an average score of 56.81% nDCG and 42.63% mAP.
Authors: Maoxun Yuan, Xingxing Wei
Object detection on visible (RGB) and infrared (IR) images, as an emerging solution to facilitate robust detection for around-the-clock applications, has received extensive attention in recent years. With the help of IR images, object detectors have been more reliable and robust in practical applications by using RGB-IR combined information. However, existing methods still suffer from modality miscalibration and fusion imprecision problems. Since transformer has the powerful capability to model the pairwise correlations between different features, in this paper, we propose a novel Calibrated and Complementary Transformer called $\mathrm{C}^2$Former to address these two problems simultaneously. In $\mathrm{C}^2$Former, we design an Inter-modality Cross-Attention (ICA) module to obtain the calibrated and complementary features by learning the cross-attention relationship between the RGB and IR modality. To reduce the computational cost caused by computing the global attention in ICA, an Adaptive Feature Sampling (AFS) module is introduced to decrease the dimension of feature maps. Because $\mathrm{C}^2$Former performs in the feature domain, it can be embedded into existed RGB-IR object detectors via the backbone network. Thus, one single-stage and one two-stage object detector both incorporating our $\mathrm{C}^2$Former are constructed to evaluate its effectiveness and versatility. With extensive experiments on the DroneVehicle and KAIST RGB-IR datasets, we verify that our method can fully utilize the RGB-IR complementary information and achieve robust detection results. The code is available at https://github.com/yuanmaoxun/Calibrated-and-Complementary-Transformer-for-RGB-Infrared-Object-Detection.git.
Authors: Yeongwoong Kim, Hyewon Jeong, Janghyun Yu, Younhee Kim, Jooyoung Lee, Se Yoon Jeong, Hui Yong Kim
The proliferation of deep learning-based machine vision applications has given rise to a new type of compression, so called video coding for machine (VCM). VCM differs from traditional video coding in that it is optimized for machine vision performance instead of human visual quality. In the feature compression track of MPEG-VCM, multi-scale features extracted from images are subject to compression. Recent feature compression works have demonstrated that the versatile video coding (VVC) standard-based approach can achieve a BD-rate reduction of up to 96% against MPEG-VCM feature anchor. However, it is still sub-optimal as VVC was not designed for extracted features but for natural images. Moreover, the high encoding complexity of VVC makes it difficult to design a lightweight encoder without sacrificing performance. To address these challenges, we propose a novel multi-scale feature compression method that enables both the end-to-end optimization on the extracted features and the design of lightweight encoders. The proposed model combines a learnable compressor with a multi-scale feature fusion network so that the redundancy in the multi-scale features is effectively removed. Instead of simply cascading the fusion network and the compression network, we integrate the fusion and encoding processes in an interleaved way. Our model first encodes a larger-scale feature to obtain a latent representation and then fuses the latent with a smaller-scale feature. This process is successively performed until the smallest-scale feature is fused and then the encoded latent at the final stage is entropy-coded for transmission. The results show that our model outperforms previous approaches by at least 52% BD-rate reduction and has $\times5$ to $\times27$ times less encoding time for object detection...
Authors: Che-Jui Chang, Honglu Zhou, Parth Goel, Aditya Bhat, Seonghyeon Moon, Samuel S. Sohn, Sejong Yoon, Vladimir Pavlovic, Mubbasir Kapadia
The understanding of complex human interactions and group activities has garnered attention in human-centric computer vision. However, the advancement of the related tasks is hindered due to the difficulty of obtaining large-scale labeled real-world datasets. To mitigate the issue, we propose M3Act, a multi-view multi-group multi-person human atomic action and group activity data generator. Powered by the Unity engine, M3Act contains simulation-ready 3D scenes and human assets, configurable lighting and camera systems, highly parameterized modular group activities, and a large degree of domain randomization during the data generation process. Our data generator is capable of generating large-scale datasets of human activities with multiple viewpoints, modalities (RGB images, 2D poses, 3D motions), and high-quality annotations for individual persons and multi-person groups (2D bounding boxes, instance segmentation masks, individual actions and group activity categories). Using M3Act, we perform synthetic data pre-training for 2D skeleton-based group activity recognition and RGB-based multi-person pose tracking. The results indicate that learning from our synthetic datasets largely improves the model performances on real-world datasets, with the highest gain of 5.59% and 7.32% respectively in group and person recognition accuracy on CAD2, as well as an improvement of 6.63 in MOTP on HiEve. Pre-training with our synthetic data also leads to faster model convergence on downstream tasks (up to 6.8% faster). Moreover, M3Act opens new research problems for 3D group activity generation. We release M3Act3D, an 87.6-hour 3D motion dataset of human activities with larger group sizes and higher complexity of inter-person interactions than previous multi-person datasets. We define multiple metrics and propose a competitive baseline for the novel task.
Authors: Seunghyeon Seo, Yeonjin Chang, Nojun Kwak
Neural Radiance Field (NeRF) has been a mainstream in novel view synthesis with its remarkable quality of rendered images and simple architecture. Although NeRF has been developed in various directions improving continuously its performance, the necessity of a dense set of multi-view images still exists as a stumbling block to progress for practical application. In this work, we propose FlipNeRF, a novel regularization method for few-shot novel view synthesis by utilizing our proposed flipped reflection rays. The flipped reflection rays are explicitly derived from the input ray directions and estimated normal vectors, and play a role of effective additional training rays while enabling to estimate more accurate surface normals and learn the 3D geometry effectively. Since the surface normal and the scene depth are both derived from the estimated densities along a ray, the accurate surface normal leads to more exact depth estimation, which is a key factor for few-shot novel view synthesis. Furthermore, with our proposed Uncertainty-aware Emptiness Loss and Bottleneck Feature Consistency Loss, FlipNeRF is able to estimate more reliable outputs with reducing floating artifacts effectively across the different scene structures, and enhance the feature-level consistency between the pair of the rays cast toward the photo-consistent pixels without any additional feature extractor, respectively. Our FlipNeRF achieves the SOTA performance on the multiple benchmarks across all the scenarios.
Authors: Jiarui Wang, Huiyu Duan, Jing Liu, Shi Chen, Xiongkuo Min, Guangtao Zhai
In this paper, in order to get a better understanding of the human visual preferences for AIGIs, a large-scale IQA database for AIGC is established, which is named as AIGCIQA2023. We first generate over 2000 images based on 6 state-of-the-art text-to-image generation models using 100 prompts. Based on these images, a well-organized subjective experiment is conducted to assess the human visual preferences for each image from three perspectives including quality, authenticity and correspondence. Finally, based on this large-scale database, we conduct a benchmark experiment to evaluate the performance of several state-of-the-art IQA metrics on our constructed database.
Authors: Shitao Tang, Fuyang Zhang, Jiacheng Chen, Peng Wang, Yasutaka Furukawa
This paper introduces MVDiffusion, a simple yet effective multi-view image generation method for scenarios where pixel-to-pixel correspondences are available, such as perspective crops from panorama or multi-view images given geometry (depth maps and poses). Unlike prior models that rely on iterative image warping and inpainting, MVDiffusion concurrently generates all images with a global awareness, encompassing high resolution and rich content, effectively addressing the error accumulation prevalent in preceding models. MVDiffusion specifically incorporates a correspondence-aware attention mechanism, enabling effective cross-view interaction. This mechanism underpins three pivotal modules: 1) a generation module that produces low-resolution images while maintaining global correspondence, 2) an interpolation module that densifies spatial coverage between images, and 3) a super-resolution module that upscales into high-resolution outputs. In terms of panoramic imagery, MVDiffusion can generate high-resolution photorealistic images up to 1024$\times$1024 pixels. For geometry-conditioned multi-view image generation, MVDiffusion demonstrates the first method capable of generating a textured map of a scene mesh. The project page is at https://mvdiffusion.github.io.
Authors: Huafeng Li, Le Xu, Yafei Zhang, Dapeng Tao, Zhengtao Yu
In visible-infrared video person re-identification (re-ID), extracting features not affected by complex scenes (such as modality, camera views, pedestrian pose, background, etc.) changes, and mining and utilizing motion information are the keys to solving cross-modal pedestrian identity matching. To this end, the paper proposes a new visible-infrared video person re-ID method from a novel perspective, i.e., adversarial self-attack defense and spatial-temporal relation mining. In this work, the changes of views, posture, background and modal discrepancy are considered as the main factors that cause the perturbations of person identity features. Such interference information contained in the training samples is used as an adversarial perturbation. It performs adversarial attacks on the re-ID model during the training to make the model more robust to these unfavorable factors. The attack from the adversarial perturbation is introduced by activating the interference information contained in the input samples without generating adversarial samples, and it can be thus called adversarial self-attack. This design allows adversarial attack and defense to be integrated into one framework. This paper further proposes a spatial-temporal information-guided feature representation network to use the information in video sequences. The network cannot only extract the information contained in the video-frame sequences but also use the relation of the local information in space to guide the network to extract more robust features. The proposed method exhibits compelling performance on large-scale cross-modality video datasets. The source code of the proposed method will be released at https://github.com/lhf12278/xxx.
Authors: Mingze Yuan, Yingda Xia, Xin Chen, Jiawen Yao, Junli Wang, Mingyan Qiu, Hexin Dong, Jingren Zhou, Bin Dong, Le Lu, Li Zhang, Zaiyi Liu, Ling Zhang
Gastric cancer is the third leading cause of cancer-related mortality worldwide, but no guideline-recommended screening test exists. Existing methods can be invasive, expensive, and lack sensitivity to identify early-stage gastric cancer. In this study, we explore the feasibility of using a deep learning approach on non-contrast CT scans for gastric cancer detection. We propose a novel cluster-induced Mask Transformer that jointly segments the tumor and classifies abnormality in a multi-task manner. Our model incorporates learnable clusters that encode the texture and shape prototypes of gastric cancer, utilizing self- and cross-attention to interact with convolutional features. In our experiments, the proposed method achieves a sensitivity of 85.0% and specificity of 92.6% for detecting gastric tumors on a hold-out test set consisting of 100 patients with cancer and 148 normal. In comparison, two radiologists have an average sensitivity of 73.5% and specificity of 84.3%. We also obtain a specificity of 97.7% on an external test set with 903 normal cases. Our approach performs comparably to established state-of-the-art gastric cancer screening tools like blood testing and endoscopy, while also being more sensitive in detecting early-stage cancer. This demonstrates the potential of our approach as a novel, non-invasive, low-cost, and accurate method for opportunistic gastric cancer screening.
Authors: Chao Wang, Zheng Tang
In the context of label-efficient learning on video data, the distillation method and the structural design of the teacher-student architecture have a significant impact on knowledge distillation. However, the relationship between these factors has been overlooked in previous research. To address this gap, we propose a new weakly supervised learning framework for knowledge distillation in video classification that is designed to improve the efficiency and accuracy of the student model. Our approach leverages the concept of substage-based learning to distill knowledge based on the combination of student substages and the correlation of corresponding substages. We also employ the progressive cascade training method to address the accuracy loss caused by the large capacity gap between the teacher and the student. Additionally, we propose a pseudo-label optimization strategy to improve the initial data label. To optimize the loss functions of different distillation substages during the training process, we introduce a new loss method based on feature distribution. We conduct extensive experiments on both real and simulated data sets, demonstrating that our proposed approach outperforms existing distillation methods in terms of knowledge distillation for video classification tasks. Our proposed substage-based distillation approach has the potential to inform future research on label-efficient learning for video data.
Authors: Jože M. Rožanec, Elias Montini, Vincenzo Cutrona, Dimitrios Papamartzivanos, Timotej Klemenčič, Blaž Fortuna, Dunja Mladenić, Entso Veliou, Thanassis Giannetsos, Christos Emmanouilidis
Industrial revolutions have historically disrupted manufacturing by introducing automation into production. Increasing automation reshapes the role of the human worker. Advances in robotics and artificial intelligence open new frontiers of human-machine collaboration. Such collaboration can be realized considering two sub-fields of artificial intelligence: active learning and explainable artificial intelligence. Active learning aims to devise strategies that help obtain data that allows machine learning algorithms to learn better. On the other hand, explainable artificial intelligence aims to make the machine learning models intelligible to the human person. The present work first describes Industry 5.0, human-machine collaboration, and state-of-the-art regarding quality inspection, emphasizing visual inspection. Then it outlines how human-machine collaboration could be realized and enhanced in visual inspection. Finally, some of the results obtained in the EU H2020 STAR project regarding visual inspection are shared, considering artificial intelligence, human digital twins, and cybersecurity.
Authors: Chantal Pellegrini, Matthias Keicher, Ege Özsoy, Nassir Navab
Radiology reporting is a crucial part of the communication between radiologists and other medical professionals, but it can be time-consuming and error-prone. One approach to alleviate this is structured reporting, which saves time and enables a more accurate evaluation than free-text reports. However, there is limited research on automating structured reporting, and no public benchmark is available for evaluating and comparing different methods. To close this gap, we introduce Rad-ReStruct, a new benchmark dataset that provides fine-grained, hierarchically ordered annotations in the form of structured reports for X-Ray images. We model the structured reporting task as hierarchical visual question answering (VQA) and propose hi-VQA, a novel method that considers prior context in the form of previously asked questions and answers for populating a structured radiology report. Our experiments show that hi-VQA achieves competitive performance to the state-of-the-art on the medical VQA benchmark VQARad while performing best among methods without domain-specific vision-language pretraining and provides a strong baseline on Rad-ReStruct. Our work represents a significant step towards the automated population of structured radiology reports and provides a valuable first benchmark for future research in this area. We will make all annotations and our code for annotation generation, model evaluation, and training publicly available upon acceptance. Our dataset and code is available at https://github.com/ChantalMP/Rad-ReStruct.
Authors: Huafeng Li, Yanmei Mao, Yafei Zhang, Guanqiu Qi, Zhengtao Yu
Existing person re-identification (re-ID) research mainly focuses on pedestrian identity matching across cameras in adjacent areas. However, in reality, it is inevitable to face the problem of pedestrian identity matching across long-distance scenes. The cross-camera pedestrian samples collected from long-distance scenes often have no positive samples. It is extremely challenging to use cross-camera negative samples to achieve cross-region pedestrian identity matching. Therefore, a novel domain-adaptive person re-ID method that focuses on cross-camera consistent discriminative feature learning under the supervision of unpaired samples is proposed. This method mainly includes category synergy co-promotion module (CSCM) and cross-camera consistent feature learning module (CCFLM). In CSCM, a task-specific feature recombination (FRT) mechanism is proposed. This mechanism first groups features according to their contributions to specific tasks. Then an interactive promotion learning (IPL) scheme between feature groups is developed and embedded in this mechanism to enhance feature discriminability. Since the control parameters of the specific task model are reduced after division by task, the generalization ability of the model is improved. In CCFLM, instance-level feature distribution alignment and cross-camera identity consistent learning methods are constructed. Therefore, the supervised model training is achieved under the style supervision of the target domain by exchanging styles between source-domain samples and target-domain samples, and the challenges caused by the lack of cross-camera paired samples are solved by utilizing cross-camera similar samples. In experiments, three challenging datasets are used as target domains, and the effectiveness of the proposed method is demonstrated through four experimental settings.
Authors: Jialiang Sun, Wen Yao, Tingsong Jiang, Xiaoqian Chen
Deep neural networks have proven to be vulnerable to adversarial attacks in the form of adding specific perturbations on images to make wrong outputs. Designing stronger adversarial attack methods can help more reliably evaluate the robustness of DNN models. To release the harbor burden and improve the attack performance, auto machine learning (AutoML) has recently emerged as one successful technique to help automatically find the near-optimal adversarial attack strategy. However, existing works about AutoML for adversarial attacks only focus on $L_{\infty}$-norm-based perturbations. In fact, semantic perturbations attract increasing attention due to their naturalnesses and physical realizability. To bridge the gap between AutoML and semantic adversarial attacks, we propose a novel method called multi-objective evolutionary search of variable-length composite semantic perturbations (MES-VCSP). Specifically, we construct the mathematical model of variable-length composite semantic perturbations, which provides five gradient-based semantic attack methods. The same type of perturbation in an attack sequence is allowed to be performed multiple times. Besides, we introduce the multi-objective evolutionary search consisting of NSGA-II and neighborhood search to find near-optimal variable-length attack sequences. Experimental results on CIFAR10 and ImageNet datasets show that compared with existing methods, MES-VCSP can obtain adversarial examples with a higher attack success rate, more naturalness, and less time cost.
Authors: Syed Talal Wasim, Muhammad Uzair Khattak, Muzammal Naseer, Salman Khan, Mubarak Shah, Fahad Shahbaz Khan
Recent video recognition models utilize Transformer models for long-range spatio-temporal context modeling. Video transformer designs are based on self-attention that can model global context at a high computational cost. In comparison, convolutional designs for videos offer an efficient alternative but lack long-range dependency modeling. Towards achieving the best of both designs, this work proposes Video-FocalNet, an effective and efficient architecture for video recognition that models both local and global contexts. Video-FocalNet is based on a spatio-temporal focal modulation architecture that reverses the interaction and aggregation steps of self-attention for better efficiency. Further, the aggregation step and the interaction step are both implemented using efficient convolution and element-wise multiplication operations that are computationally less expensive than their self-attention counterparts on video representations. We extensively explore the design space of focal modulation-based spatio-temporal context modeling and demonstrate our parallel spatial and temporal encoding design to be the optimal choice. Video-FocalNets perform favorably well against the state-of-the-art transformer-based models for video recognition on three large-scale datasets (Kinetics-400, Kinetics-600, and SS-v2) at a lower computational cost. Our code/models are released at https://github.com/TalalWasim/Video-FocalNets.
Authors: Xiaoyan Yang, Dingbo Lu, Yang Li, Chenhui Li, Changbo Wang
In recent years, novel view synthesis has gained popularity in generating high-fidelity images. While demonstrating superior performance in the task of synthesizing novel views, the majority of these methods are still based on the conventional multi-layer perceptron for scene embedding. Furthermore, light field models suffer from geometric blurring during pixel rendering, while radiance field-based volume rendering methods have multiple solutions for a certain target of density distribution integration. To address these issues, we introduce the Convolutional Neural Radiance Fields to model the derivatives of radiance along rays. Based on 1D convolutional operations, our proposed method effectively extracts potential ray representations through a structured neural network architecture. Besides, with the proposed ray modeling, a proposed recurrent module is employed to solve geometric ambiguity in the fully neural rendering process. Extensive experiments demonstrate the promising results of our proposed model compared with existing state-of-the-art methods.
Authors: Xiaofei Chen, Yuting He, Cheng Xue, Rongjun Ge, Shuo Li, Guanyu Yang
The foundation models based on pre-training technology have significantly advanced artificial intelligence from theoretical to practical applications. These models have facilitated the feasibility of computer-aided diagnosis for widespread use. Medical contrastive vision-language pre-training, which does not require human annotations, is an effective approach for guiding representation learning using description information in diagnostic reports. However, the effectiveness of pre-training is limited by the large-scale semantic overlap and shifting problems in medical field. To address these issues, we propose the Knowledge-Boosting Contrastive Vision-Language Pre-training framework (KoBo), which integrates clinical knowledge into the learning of vision-language semantic consistency. The framework uses an unbiased, open-set sample-wise knowledge representation to measure negative sample noise and supplement the correspondence between vision-language mutual information and clinical knowledge. Extensive experiments validate the effect of our framework on eight tasks including classification, segmentation, retrieval, and semantic relatedness, achieving comparable or better performance with the zero-shot or few-shot settings. Our code is open on https://github.com/ChenXiaoFei-CS/KoBo.