Pub Date : 2022-06-01DOI: 10.1109/CVPR52688.2022.00962
Arnav Chavan, Rishabh Tiwari, Udbhav Bamba, D. Gupta
Gradient based meta-learning methods are prone to overfit on the meta-training set, and this behaviour is more prominent with large and complex networks. Moreover, large networks restrict the application of meta-learning models on low-power edge devices. While choosing smaller networks avoid these issues to a certain extent, it affects the overall generalization leading to reduced performance. Clearly, there is an approximately optimal choice of network architecture that is best suited for every meta-learning problem, however, identifying it beforehand is not straight-forward. In this paper, we present Metadock, a task-specific dynamic kernel selection strategy for designing compressed CNN models that generalize well on unseen tasks in meta-learning. Our method is based on the hypothesis that for a given set of similar tasks, not all kernels of the network are needed by each individual task. Rather, each task uses only a fraction of the kernels, and the selection of the kernels per task can be learnt dynamically as a part of the inner update steps. Metadockcompresses the meta-model as well as the task-specific inner models, thus providing significant reduction in model size for each task, and through constraining the number of active kernels for every task, it implicitly mitigates the issue of meta-overfitting. We show that for the same inference budget, pruned versions of large CNN models obtained using our approach consistently outperform the conventional choices of CNN models. Metadock couples well with popular meta-learning approaches such as iMAML [22]. The efficacy of our method is validated on CIFAR-fs [1] and mini-ImageNet [28] datasets, and we have observed that our approach can provide improvements in model accuracy of up to 2% on standard meta-learning benchmark, while reducing the model size by more than 75%. Our code is available at https://github.com/transmuteAI/MetaDOCK.
{"title":"Dynamic Kernel Selection for Improved Generalization and Memory Efficiency in Meta-learning","authors":"Arnav Chavan, Rishabh Tiwari, Udbhav Bamba, D. Gupta","doi":"10.1109/CVPR52688.2022.00962","DOIUrl":"https://doi.org/10.1109/CVPR52688.2022.00962","url":null,"abstract":"Gradient based meta-learning methods are prone to overfit on the meta-training set, and this behaviour is more prominent with large and complex networks. Moreover, large networks restrict the application of meta-learning models on low-power edge devices. While choosing smaller networks avoid these issues to a certain extent, it affects the overall generalization leading to reduced performance. Clearly, there is an approximately optimal choice of network architecture that is best suited for every meta-learning problem, however, identifying it beforehand is not straight-forward. In this paper, we present Metadock, a task-specific dynamic kernel selection strategy for designing compressed CNN models that generalize well on unseen tasks in meta-learning. Our method is based on the hypothesis that for a given set of similar tasks, not all kernels of the network are needed by each individual task. Rather, each task uses only a fraction of the kernels, and the selection of the kernels per task can be learnt dynamically as a part of the inner update steps. Metadockcompresses the meta-model as well as the task-specific inner models, thus providing significant reduction in model size for each task, and through constraining the number of active kernels for every task, it implicitly mitigates the issue of meta-overfitting. We show that for the same inference budget, pruned versions of large CNN models obtained using our approach consistently outperform the conventional choices of CNN models. Metadock couples well with popular meta-learning approaches such as iMAML [22]. The efficacy of our method is validated on CIFAR-fs [1] and mini-ImageNet [28] datasets, and we have observed that our approach can provide improvements in model accuracy of up to 2% on standard meta-learning benchmark, while reducing the model size by more than 75%. Our code is available at https://github.com/transmuteAI/MetaDOCK.","PeriodicalId":355552,"journal":{"name":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134050051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-01DOI: 10.1109/CVPR52688.2022.00696
Xu Yao, Yang Bai, Xinyun Zhang, Yuechen Zhang, Qi Sun, Ran Chen, Ruiyu Li, Bei Yu
Domain generalization refers to the problem of training a model from a collection of different source domains that can directly generalize to the unseen target domains. A promising solution is contrastive learning, which attempts to learn domain-invariant representations by exploiting rich semantic relations among sample-to-sample pairs from different domains. A simple approach is to pull positive sample pairs from different domains closer while pushing other negative pairs further apart. In this paper, we find that directly applying contrastive-based methods (e.g., supervised contrastive learning) are not effective in domain generalization. We argue that aligning positive sample-to-sample pairs tends to hinder the model generalization due to the significant distribution gaps between different domains. To address this issue, we propose a novel proxy-based contrastive learning method, which replaces the original sample-to-sample relations with proxy-to-sample relations, significantly alleviating the positive alignment issue. Experiments on the four standard benchmarks demonstrate the effectiveness of the proposed method. Furthermore, we also consider a more complex scenario where no ImageNet pre-trained models are provided. Our method consistently shows better performance.
{"title":"PCL: Proxy-based Contrastive Learning for Domain Generalization","authors":"Xu Yao, Yang Bai, Xinyun Zhang, Yuechen Zhang, Qi Sun, Ran Chen, Ruiyu Li, Bei Yu","doi":"10.1109/CVPR52688.2022.00696","DOIUrl":"https://doi.org/10.1109/CVPR52688.2022.00696","url":null,"abstract":"Domain generalization refers to the problem of training a model from a collection of different source domains that can directly generalize to the unseen target domains. A promising solution is contrastive learning, which attempts to learn domain-invariant representations by exploiting rich semantic relations among sample-to-sample pairs from different domains. A simple approach is to pull positive sample pairs from different domains closer while pushing other negative pairs further apart. In this paper, we find that directly applying contrastive-based methods (e.g., supervised contrastive learning) are not effective in domain generalization. We argue that aligning positive sample-to-sample pairs tends to hinder the model generalization due to the significant distribution gaps between different domains. To address this issue, we propose a novel proxy-based contrastive learning method, which replaces the original sample-to-sample relations with proxy-to-sample relations, significantly alleviating the positive alignment issue. Experiments on the four standard benchmarks demonstrate the effectiveness of the proposed method. Furthermore, we also consider a more complex scenario where no ImageNet pre-trained models are provided. Our method consistently shows better performance.","PeriodicalId":355552,"journal":{"name":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131889196","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-01DOI: 10.1109/CVPR52688.2022.01477
Qiuling Xu, Guanhong Tao, Xiangyu Zhang
We propose a novel adversarial attack targeting content features in some deep layer, that is, individual neurons in the layer. A naive method that enforces a fixed value/percentage bound for neuron activation values can hardly work and generates very noisy samples. The reason is that the level of perceptual variation entailed by a fixed value bound is non-uniform across neurons and even for the same neuron. We hence propose a novel distribution quantile bound for activation values and a polynomial barrier loss function. Given a benign input, a fixed quantile bound is translated to many value bounds, one for each neuron, based on the distributions of the neuron's activations and the current activation value on the given input. These individualized bounds enable fine-grained regulation, allowing content feature mutations with bounded perceptional variations. Our evaluation on ImageNet and five different model architectures demonstrates that our attack is effective. Compared to seven other latest adversarial attacks in both the pixel space and the feature space, our attack can achieve the state-of-the-art trade-off between attack success rate and imperceptibility. 11Code and Samples are available on Github [37].
{"title":"Bounded Adversarial Attack on Deep Content Features","authors":"Qiuling Xu, Guanhong Tao, Xiangyu Zhang","doi":"10.1109/CVPR52688.2022.01477","DOIUrl":"https://doi.org/10.1109/CVPR52688.2022.01477","url":null,"abstract":"We propose a novel adversarial attack targeting content features in some deep layer, that is, individual neurons in the layer. A naive method that enforces a fixed value/percentage bound for neuron activation values can hardly work and generates very noisy samples. The reason is that the level of perceptual variation entailed by a fixed value bound is non-uniform across neurons and even for the same neuron. We hence propose a novel distribution quantile bound for activation values and a polynomial barrier loss function. Given a benign input, a fixed quantile bound is translated to many value bounds, one for each neuron, based on the distributions of the neuron's activations and the current activation value on the given input. These individualized bounds enable fine-grained regulation, allowing content feature mutations with bounded perceptional variations. Our evaluation on ImageNet and five different model architectures demonstrates that our attack is effective. Compared to seven other latest adversarial attacks in both the pixel space and the feature space, our attack can achieve the state-of-the-art trade-off between attack success rate and imperceptibility. 11Code and Samples are available on Github [37].","PeriodicalId":355552,"journal":{"name":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127595030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
It is challenging to accurately detect camouflaged objects from their highly similar surroundings. Existing methods mainly leverage a single-stage detection fashion, while neglecting small objects with low-resolution fine edges requires more operations than the larger ones. To tackle camouflaged object detection (COD), we are inspired by humans attention coupled with the coarse-to-fine detection strategy, and thereby propose an iterative refinement framework, coined SegMaR, which integrates Segment, Magnify and Reiterate in a multi-stage detection fashion. Specifically, we design a new discriminative mask which makes the model attend on the fixation and edge regions. In addition, we leverage an attention-based sampler to magnify the object region progressively with no need of enlarging the image size. Extensive experiments show our SegMaR achieves remarkable and consistent improvements over other state-of-the-art methods. Especially, we surpass two competitive methods 7.4% and 20.0% respectively in average over standard evaluation metrics on small camouflaged objects. Additional studies provide more promising insights into Seg-MaR, including its effectiveness on the discriminative mask and its generalization to other network architectures. Code is available at https://github.com/dlut-dimt/SegMaR.
{"title":"Segment, Magnify and Reiterate: Detecting Camouflaged Objects the Hard Way","authors":"Qi Jia, Shuilian Yao, Yu Liu, Xin Fan, Risheng Liu, Zhongxuan Luo","doi":"10.1109/CVPR52688.2022.00467","DOIUrl":"https://doi.org/10.1109/CVPR52688.2022.00467","url":null,"abstract":"It is challenging to accurately detect camouflaged objects from their highly similar surroundings. Existing methods mainly leverage a single-stage detection fashion, while neglecting small objects with low-resolution fine edges requires more operations than the larger ones. To tackle camouflaged object detection (COD), we are inspired by humans attention coupled with the coarse-to-fine detection strategy, and thereby propose an iterative refinement framework, coined SegMaR, which integrates Segment, Magnify and Reiterate in a multi-stage detection fashion. Specifically, we design a new discriminative mask which makes the model attend on the fixation and edge regions. In addition, we leverage an attention-based sampler to magnify the object region progressively with no need of enlarging the image size. Extensive experiments show our SegMaR achieves remarkable and consistent improvements over other state-of-the-art methods. Especially, we surpass two competitive methods 7.4% and 20.0% respectively in average over standard evaluation metrics on small camouflaged objects. Additional studies provide more promising insights into Seg-MaR, including its effectiveness on the discriminative mask and its generalization to other network architectures. Code is available at https://github.com/dlut-dimt/SegMaR.","PeriodicalId":355552,"journal":{"name":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126572903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-01DOI: 10.1109/CVPR52688.2022.00330
Dan Liu, Libo Zhang, Yanjun Wu
Gesture recognition plays an important role in natural human-computer interaction and sign language recognition. Existing research on gesture recognition is limited to close-range interaction such as vehicle gesture control and face-to-face communication. To apply gesture recognition to long-distance interactive scenes such as meetings and smart homes, a large RGB-D video dataset LD-ConGR is established in this paper. LD-ConGR is distinguished from existing gesture datasets by its long-distance gesture collection, fine-grained annotations, and high video qual-ity. Specifically, 1) the farthest gesture provided by the LD-ConGR is captured 4m away from the camera while existing gesture datasets collect gestures within 1m from the camera; 2) besides the gesture category, the temporal segmentation of gestures and hand location are also anno-tated in LD-ConGR; 3) videos are captured at high reso-lution (1280 x 720 for color streams and 640 x 576 for depth streams) and high frame rate (30 fps). On top of the LD-ConGR, a series of experimental and studies are conducted, and the proposed gesture region estimation and key frame sampling strategies are demonstrated to be effective in dealing with long-distance gesture recognition and the uncertainty of gesture duration. The dataset and experimen-tal results presented in this paper are expected to boost the research of long-distance gesture recognition. The dataset is available at https://github.com/Diananini/LD-ConGR-CVPR2022.
手势识别在自然人机交互和手语识别中占有重要地位。现有的手势识别研究仅限于近距离互动，如车辆手势控制和面对面交流。为了将手势识别应用于会议、智能家居等远距离交互场景，本文建立了大型RGB-D视频数据集ld - conr。ld - conr以其远距离的手势采集、细粒度的注释和高视频质量区别于现有的手势数据集。具体来说，1)ld - conr提供的最远的手势是在距离相机4m的地方捕获的，而现有的手势数据集收集的是距离相机1m以内的手势;2) ld - conr除对手势类别进行标注外，还对手势的时间分割和手部位置进行标注;3)视频以高分辨率(1280 x 720彩色流和640 x 576深度流)和高帧率(30 fps)捕获。在LD-ConGR的基础上，进行了一系列的实验和研究，证明了所提出的手势区域估计和关键帧采样策略在处理远距离手势识别和手势持续时间的不确定性方面是有效的。本文的数据集和实验结果有望推动远距离手势识别的研究。该数据集可在https://github.com/Diananini/LD-ConGR-CVPR2022上获得。
{"title":"LD-ConGR: A Large RGB-D Video Dataset for Long-Distance Continuous Gesture Recognition","authors":"Dan Liu, Libo Zhang, Yanjun Wu","doi":"10.1109/CVPR52688.2022.00330","DOIUrl":"https://doi.org/10.1109/CVPR52688.2022.00330","url":null,"abstract":"Gesture recognition plays an important role in natural human-computer interaction and sign language recognition. Existing research on gesture recognition is limited to close-range interaction such as vehicle gesture control and face-to-face communication. To apply gesture recognition to long-distance interactive scenes such as meetings and smart homes, a large RGB-D video dataset LD-ConGR is established in this paper. LD-ConGR is distinguished from existing gesture datasets by its long-distance gesture collection, fine-grained annotations, and high video qual-ity. Specifically, 1) the farthest gesture provided by the LD-ConGR is captured 4m away from the camera while existing gesture datasets collect gestures within 1m from the camera; 2) besides the gesture category, the temporal segmentation of gestures and hand location are also anno-tated in LD-ConGR; 3) videos are captured at high reso-lution (1280 x 720 for color streams and 640 x 576 for depth streams) and high frame rate (30 fps). On top of the LD-ConGR, a series of experimental and studies are conducted, and the proposed gesture region estimation and key frame sampling strategies are demonstrated to be effective in dealing with long-distance gesture recognition and the uncertainty of gesture duration. The dataset and experimen-tal results presented in this paper are expected to boost the research of long-distance gesture recognition. The dataset is available at https://github.com/Diananini/LD-ConGR-CVPR2022.","PeriodicalId":355552,"journal":{"name":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131249940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The huge burden of computation and memory are two obstacles in ultra-high resolution image segmentation. To tackle these issues, most of the previous works follow the global-local refinement pipeline, which pays more attention to the memory consumption but neglects the inference speed. In comparison to the pipeline that partitions the large image into small local regions, we focus on inferring the whole image directly. In this paper, we propose ISDNet, a novel ultra-high resolution segmentation framework that integrates the shallow and deep networks in a new manner, which significantly accelerates the inference speed while achieving accurate segmentation. To further exploit the relationship between the shallow and deep features, we propose a novel Relational-Aware feature Fusion module, which ensures high performance and robustness of our framework. Extensive experiments on Deepglobe, Inria Aerial, and Cityscapes datasets demonstrate our performance is consistently superior to state-of-the-arts. Specifically, it achieves 73.30 mIoU with a speed of 27.70 FPS on Deepglobe, which is more accurate and 172 × faster than the recent competitor. Code available at https://github.com/cedricgsh/ISDNet.
{"title":"ISDNet: Integrating Shallow and Deep Networks for Efficient Ultra-high Resolution Segmentation","authors":"Shaohua Guo, Liang Liu, Zhenye Gan, Yabiao Wang, Wuhao Zhang, Chengjie Wang, Guannan Jiang, Wei Zhang, Ran Yi, Lizhuang Ma, Ke Xu","doi":"10.1109/CVPR52688.2022.00432","DOIUrl":"https://doi.org/10.1109/CVPR52688.2022.00432","url":null,"abstract":"The huge burden of computation and memory are two obstacles in ultra-high resolution image segmentation. To tackle these issues, most of the previous works follow the global-local refinement pipeline, which pays more attention to the memory consumption but neglects the inference speed. In comparison to the pipeline that partitions the large image into small local regions, we focus on inferring the whole image directly. In this paper, we propose ISDNet, a novel ultra-high resolution segmentation framework that integrates the shallow and deep networks in a new manner, which significantly accelerates the inference speed while achieving accurate segmentation. To further exploit the relationship between the shallow and deep features, we propose a novel Relational-Aware feature Fusion module, which ensures high performance and robustness of our framework. Extensive experiments on Deepglobe, Inria Aerial, and Cityscapes datasets demonstrate our performance is consistently superior to state-of-the-arts. Specifically, it achieves 73.30 mIoU with a speed of 27.70 FPS on Deepglobe, which is more accurate and 172 × faster than the recent competitor. Code available at https://github.com/cedricgsh/ISDNet.","PeriodicalId":355552,"journal":{"name":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130839521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-01DOI: 10.1109/CVPR52688.2022.02068
Tao Sun, Mattia Segu, Janis Postels, Yuxuan Wang, L. Gool, B. Schiele, F. Tombari, F. Yu
Adapting to a continuously evolving environment is a safety-critical challenge inevitably faced by all autonomous-driving systems. Existing image- and video-based driving datasets, however, fall short of capturing the mutable nature of the real world. In this paper, we introduce the largest multi-task synthetic dataset for autonomous driving, SHIFT. It presents discrete and continuous shifts in cloudiness, rain and fog intensity, time of day, and vehicle and pedestrian density. Featuring a comprehensive sensor suite and annotations for several mainstream perception tasks, SHIFT allows to investigate how a perception systems' performance degrades at increasing levels of domain shift, fostering the development of continuous adaptation strategies to mitigate this problem and assessing the robustness and generality of a model. Our dataset and benchmark toolkit are publicly available at www.vis.xyz/shift.
{"title":"SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptation","authors":"Tao Sun, Mattia Segu, Janis Postels, Yuxuan Wang, L. Gool, B. Schiele, F. Tombari, F. Yu","doi":"10.1109/CVPR52688.2022.02068","DOIUrl":"https://doi.org/10.1109/CVPR52688.2022.02068","url":null,"abstract":"Adapting to a continuously evolving environment is a safety-critical challenge inevitably faced by all autonomous-driving systems. Existing image- and video-based driving datasets, however, fall short of capturing the mutable nature of the real world. In this paper, we introduce the largest multi-task synthetic dataset for autonomous driving, SHIFT. It presents discrete and continuous shifts in cloudiness, rain and fog intensity, time of day, and vehicle and pedestrian density. Featuring a comprehensive sensor suite and annotations for several mainstream perception tasks, SHIFT allows to investigate how a perception systems' performance degrades at increasing levels of domain shift, fostering the development of continuous adaptation strategies to mitigate this problem and assessing the robustness and generality of a model. Our dataset and benchmark toolkit are publicly available at www.vis.xyz/shift.","PeriodicalId":355552,"journal":{"name":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132819592","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-01DOI: 10.1109/CVPR52688.2022.01080
Jia Gong, Zhipeng Fan, Qiuhong Ke, Hossein Rahmani, J. Liu
The existing pose estimation approaches often require a large number of annotated images to attain good estimation performance, which are laborious to acquire. To reduce the human efforts on pose annotations, we propose a novel Meta Agent Teaming Active Learning (MATAL) framework to actively select and label informative images for effective learning. Our MATAL formulates the image selection procedure as a Markov Decision Process and learns an optimal sampling policy that directly maximizes the performance of the pose estimator based on the reward. Our framework consists of a novel state-action representation as well as a multi-agent team to enable batch sampling in the active learning procedure. The framework could be effectively optimized via Meta-Optimization to accelerate the adaptation to the gradually expanded labeled data during deployment. Finally, we show experimental results on both human hand and body pose estimation benchmark datasets and demonstrate that our method significantly outperforms all baselines continuously under the same amount of annotation budget. Moreover, to obtain similar pose estimation accuracy, our MATAL framework can save around 40% labeling efforts on average compared to state-of-the-art active learning frameworks.
{"title":"Meta Agent Teaming Active Learning for Pose Estimation","authors":"Jia Gong, Zhipeng Fan, Qiuhong Ke, Hossein Rahmani, J. Liu","doi":"10.1109/CVPR52688.2022.01080","DOIUrl":"https://doi.org/10.1109/CVPR52688.2022.01080","url":null,"abstract":"The existing pose estimation approaches often require a large number of annotated images to attain good estimation performance, which are laborious to acquire. To reduce the human efforts on pose annotations, we propose a novel Meta Agent Teaming Active Learning (MATAL) framework to actively select and label informative images for effective learning. Our MATAL formulates the image selection procedure as a Markov Decision Process and learns an optimal sampling policy that directly maximizes the performance of the pose estimator based on the reward. Our framework consists of a novel state-action representation as well as a multi-agent team to enable batch sampling in the active learning procedure. The framework could be effectively optimized via Meta-Optimization to accelerate the adaptation to the gradually expanded labeled data during deployment. Finally, we show experimental results on both human hand and body pose estimation benchmark datasets and demonstrate that our method significantly outperforms all baselines continuously under the same amount of annotation budget. Moreover, to obtain similar pose estimation accuracy, our MATAL framework can save around 40% labeling efforts on average compared to state-of-the-art active learning frameworks.","PeriodicalId":355552,"journal":{"name":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132751443","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Thanks for the cross-modal retrieval techniques, visible-infrared (RGB-IR) person re-identification (Re-ID) is achieved by projecting them into a common space, allowing person Re-ID in 24-hour surveillance systems. However, with respect to the probe-to- gallery, almost all existing RGB-IR based cross-modal person Re-ID methods focus on image-to-image matching, while the video-to-video matching which contains much richer spatial- and temporal-information remains under-explored. In this paper, we primarily study the video-based cross-modal per-son Re-ID method. To achieve this task, a video-based RGB-IR dataset is constructed, in which 927 valid identities with 463,259 frames and 21,863 tracklets captured by 12 RGB/IR cameras are collected. Based on our constructed dataset, we prove that with the increase of frames in a tracklet, the performance does meet more enhancement, demonstrating the significance of video-to-video matching in RGB-IR person Re-ID. Additionally, a novel method is further proposed, which not only projects two modalities to a modal-invariant subspace, but also extracts the temporal-memory for motion-invariant. Thanks to these two strategies, much better results are achieved on our video-based cross-modal person Re-ID. The code and dataset are released at: https://github.com/VCM-project233/MITML.
{"title":"Learning Modal-Invariant and Temporal-Memory for Video-based Visible-Infrared Person Re-Identification","authors":"Xinyu Lin, Jinxing Li, Zeyu Ma, Huafeng Li, Shuang Li, Kaixiong Xu, Guangming Lu, Dafan Zhang","doi":"10.1109/CVPR52688.2022.02030","DOIUrl":"https://doi.org/10.1109/CVPR52688.2022.02030","url":null,"abstract":"Thanks for the cross-modal retrieval techniques, visible-infrared (RGB-IR) person re-identification (Re-ID) is achieved by projecting them into a common space, allowing person Re-ID in 24-hour surveillance systems. However, with respect to the probe-to- gallery, almost all existing RGB-IR based cross-modal person Re-ID methods focus on image-to-image matching, while the video-to-video matching which contains much richer spatial- and temporal-information remains under-explored. In this paper, we primarily study the video-based cross-modal per-son Re-ID method. To achieve this task, a video-based RGB-IR dataset is constructed, in which 927 valid identities with 463,259 frames and 21,863 tracklets captured by 12 RGB/IR cameras are collected. Based on our constructed dataset, we prove that with the increase of frames in a tracklet, the performance does meet more enhancement, demonstrating the significance of video-to-video matching in RGB-IR person Re-ID. Additionally, a novel method is further proposed, which not only projects two modalities to a modal-invariant subspace, but also extracts the temporal-memory for motion-invariant. Thanks to these two strategies, much better results are achieved on our video-based cross-modal person Re-ID. The code and dataset are released at: https://github.com/VCM-project233/MITML.","PeriodicalId":355552,"journal":{"name":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133131540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-01DOI: 10.1109/CVPR52688.2022.00080
Vinit Veerendraveer Singh, C. Kambhamettu
Data augmentations are commonly used to increase the robustness of deep neural networks. In most contemporary research, the networks do not decide the augmentations; they are task-agnostic, and grid search determines their magnitudes. Furthermore, augmentations applicable to lower-dimensional data do not easily extend to higher-dimensional data and vice versa. This paper presents an auto-augmenter for images and meshes (AIM) that easily incorporates into neural networks at training and inference times. It Jointly optimizes with the network to produce constrained, non-rigid deformations in the data. AIM predicts sample-aware deformations suited for a task, and our experiments confirm its effectiveness with various networks.
{"title":"AIM: an Auto-Augmenter for Images and Meshes","authors":"Vinit Veerendraveer Singh, C. Kambhamettu","doi":"10.1109/CVPR52688.2022.00080","DOIUrl":"https://doi.org/10.1109/CVPR52688.2022.00080","url":null,"abstract":"Data augmentations are commonly used to increase the robustness of deep neural networks. In most contemporary research, the networks do not decide the augmentations; they are task-agnostic, and grid search determines their magnitudes. Furthermore, augmentations applicable to lower-dimensional data do not easily extend to higher-dimensional data and vice versa. This paper presents an auto-augmenter for images and meshes (AIM) that easily incorporates into neural networks at training and inference times. It Jointly optimizes with the network to produce constrained, non-rigid deformations in the data. AIM predicts sample-aware deformations suited for a task, and our experiments confirm its effectiveness with various networks.","PeriodicalId":355552,"journal":{"name":"2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128866139","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：