Pub Date : 2020-06-01DOI: 10.1109/cvpr42600.2020.00476
Xiaoheng Jiang, Li Zhang, Mingliang Xu, Tianzhu Zhang, Pei Lv, Bing Zhou, Xin Yang, Yanwei Pang
Convolutional Neural Network (CNN) based methods generally take crowd counting as a regression task by outputting crowd densities. They learn the mapping between image contents and crowd density distributions. Though having achieved promising results, these data-driven counting networks are prone to overestimate or underestimate people counts of regions with different density patterns, which degrades the whole count accuracy. To overcome this problem, we propose an approach to alleviate the counting performance differences in different regions. Specifically, our approach consists of two networks named Density Attention Network (DANet) and Attention Scaling Network (ASNet). DANet provides ASNet with attention masks related to regions of different density levels. ASNet first generates density maps and scaling factors and then multiplies them by attention masks to output separate attention-based density maps. These density maps are summed to give the final density map. The attention scaling factors help attenuate the estimation errors in different regions. Furthermore, we present a novel Adaptive Pyramid Loss (APLoss) to hierarchically calculate the estimation losses of sub-regions, which alleviates the training bias. Extensive experiments on four challenging datasets (ShanghaiTech Part A, UCF_CC_50, UCF-QNRF, and WorldExpo'10) demonstrate the superiority of the proposed approach.
{"title":"Attention Scaling for Crowd Counting","authors":"Xiaoheng Jiang, Li Zhang, Mingliang Xu, Tianzhu Zhang, Pei Lv, Bing Zhou, Xin Yang, Yanwei Pang","doi":"10.1109/cvpr42600.2020.00476","DOIUrl":"https://doi.org/10.1109/cvpr42600.2020.00476","url":null,"abstract":"Convolutional Neural Network (CNN) based methods generally take crowd counting as a regression task by outputting crowd densities. They learn the mapping between image contents and crowd density distributions. Though having achieved promising results, these data-driven counting networks are prone to overestimate or underestimate people counts of regions with different density patterns, which degrades the whole count accuracy. To overcome this problem, we propose an approach to alleviate the counting performance differences in different regions. Specifically, our approach consists of two networks named Density Attention Network (DANet) and Attention Scaling Network (ASNet). DANet provides ASNet with attention masks related to regions of different density levels. ASNet first generates density maps and scaling factors and then multiplies them by attention masks to output separate attention-based density maps. These density maps are summed to give the final density map. The attention scaling factors help attenuate the estimation errors in different regions. Furthermore, we present a novel Adaptive Pyramid Loss (APLoss) to hierarchically calculate the estimation losses of sub-regions, which alleviates the training bias. Extensive experiments on four challenging datasets (ShanghaiTech Part A, UCF_CC_50, UCF-QNRF, and WorldExpo'10) demonstrate the superiority of the proposed approach.","PeriodicalId":6715,"journal":{"name":"2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"165 1","pages":"4705-4714"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76752207","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 : 2020-06-01DOI: 10.1109/cvpr42600.2020.01404
Yifei Huang, Yusuke Sugano, Yoichi Sato
Temporal relations among multiple action segments play an important role in action segmentation especially when observations are limited (e.g., actions are occluded by other objects or happen outside a field of view). In this paper, we propose a network module called Graph-based Temporal Reasoning Module (GTRM) that can be built on top of existing action segmentation models to learn the relation of multiple action segments in various time spans. We model the relations by using two Graph Convolution Networks (GCNs) where each node represents an action segment. The two graphs have different edge properties to account for boundary regression and classification tasks, respectively. By applying graph convolution, we can update each node's representation based on its relation with neighboring nodes. The updated representation is then used for improved action segmentation. We evaluate our model on the challenging egocentric datasets namely EGTEA and EPIC-Kitchens, where actions may be partially observed due to the viewpoint restriction. The results show that our proposed GTRM outperforms state-of-the-art action segmentation models by a large margin. We also demonstrate the effectiveness of our model on two third-person video datasets, the 50Salads dataset and the Breakfast dataset.
{"title":"Improving Action Segmentation via Graph-Based Temporal Reasoning","authors":"Yifei Huang, Yusuke Sugano, Yoichi Sato","doi":"10.1109/cvpr42600.2020.01404","DOIUrl":"https://doi.org/10.1109/cvpr42600.2020.01404","url":null,"abstract":"Temporal relations among multiple action segments play an important role in action segmentation especially when observations are limited (e.g., actions are occluded by other objects or happen outside a field of view). In this paper, we propose a network module called Graph-based Temporal Reasoning Module (GTRM) that can be built on top of existing action segmentation models to learn the relation of multiple action segments in various time spans. We model the relations by using two Graph Convolution Networks (GCNs) where each node represents an action segment. The two graphs have different edge properties to account for boundary regression and classification tasks, respectively. By applying graph convolution, we can update each node's representation based on its relation with neighboring nodes. The updated representation is then used for improved action segmentation. We evaluate our model on the challenging egocentric datasets namely EGTEA and EPIC-Kitchens, where actions may be partially observed due to the viewpoint restriction. The results show that our proposed GTRM outperforms state-of-the-art action segmentation models by a large margin. We also demonstrate the effectiveness of our model on two third-person video datasets, the 50Salads dataset and the Breakfast dataset.","PeriodicalId":6715,"journal":{"name":"2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"61 1","pages":"14021-14031"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78121801","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 : 2020-06-01DOI: 10.1109/CVPR42600.2020.01095
Xiaoyan Wei, Tianzhu Zhang, Yan Li, Yongdong Zhang, Feng Wu
The key of image and sentence matching is to accurately measure the visual-semantic similarity between an image and a sentence. However, most existing methods make use of only the intra-modality relationship within each modality or the inter-modality relationship between image regions and sentence words for the cross-modal matching task. Different from them, in this work, we propose a novel MultiModality Cross Attention (MMCA) Network for image and sentence matching by jointly modeling the intra-modality and inter-modality relationships of image regions and sentence words in a unified deep model. In the proposed MMCA, we design a novel cross-attention mechanism, which is able to exploit not only the intra-modality relationship within each modality, but also the inter-modality relationship between image regions and sentence words to complement and enhance each other for image and sentence matching. Extensive experimental results on two standard benchmarks including Flickr30K and MS-COCO demonstrate that the proposed model performs favorably against state-of-the-art image and sentence matching methods.
{"title":"Multi-Modality Cross Attention Network for Image and Sentence Matching","authors":"Xiaoyan Wei, Tianzhu Zhang, Yan Li, Yongdong Zhang, Feng Wu","doi":"10.1109/CVPR42600.2020.01095","DOIUrl":"https://doi.org/10.1109/CVPR42600.2020.01095","url":null,"abstract":"The key of image and sentence matching is to accurately measure the visual-semantic similarity between an image and a sentence. However, most existing methods make use of only the intra-modality relationship within each modality or the inter-modality relationship between image regions and sentence words for the cross-modal matching task. Different from them, in this work, we propose a novel MultiModality Cross Attention (MMCA) Network for image and sentence matching by jointly modeling the intra-modality and inter-modality relationships of image regions and sentence words in a unified deep model. In the proposed MMCA, we design a novel cross-attention mechanism, which is able to exploit not only the intra-modality relationship within each modality, but also the inter-modality relationship between image regions and sentence words to complement and enhance each other for image and sentence matching. Extensive experimental results on two standard benchmarks including Flickr30K and MS-COCO demonstrate that the proposed model performs favorably against state-of-the-art image and sentence matching methods.","PeriodicalId":6715,"journal":{"name":"2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"33 1","pages":"10938-10947"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78215716","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 : 2020-06-01DOI: 10.1109/cvpr42600.2020.00511
Ruixu Liu, Ju Shen, He Wang, Chen Chen, S. Cheung, V. Asari
We propose a novel attention-based framework for 3D human pose estimation from a monocular video. Despite the general success of end-to-end deep learning paradigms, our approach is based on two key observations: (1) temporal incoherence and jitter are often yielded from a single frame prediction; (2) error rate can be remarkably reduced by increasing the receptive field in a video. Therefore, we design an attentional mechanism to adaptively identify significant frames and tensor outputs from each deep neural net layer, leading to a more optimal estimation. To achieve large temporal receptive fields, multi-scale dilated convolutions are employed to model long-range dependencies among frames. The architecture is straightforward to implement and can be flexibly adopted for real-time applications. Any off-the-shelf 2D pose estimation system, e.g. Mocap libraries, can be easily integrated in an ad-hoc fashion. We both quantitatively and qualitatively evaluate our method on various standard benchmark datasets (e.g. Human3.6M, HumanEva). Our method considerably outperforms all the state-of-the-art algorithms up to 8% error reduction (average mean per joint position error: 34.7) as compared to the best-reported results. Code is available at: (https://github.com/lrxjason/Attention3DHumanPose)
{"title":"Attention Mechanism Exploits Temporal Contexts: Real-Time 3D Human Pose Reconstruction","authors":"Ruixu Liu, Ju Shen, He Wang, Chen Chen, S. Cheung, V. Asari","doi":"10.1109/cvpr42600.2020.00511","DOIUrl":"https://doi.org/10.1109/cvpr42600.2020.00511","url":null,"abstract":"We propose a novel attention-based framework for 3D human pose estimation from a monocular video. Despite the general success of end-to-end deep learning paradigms, our approach is based on two key observations: (1) temporal incoherence and jitter are often yielded from a single frame prediction; (2) error rate can be remarkably reduced by increasing the receptive field in a video. Therefore, we design an attentional mechanism to adaptively identify significant frames and tensor outputs from each deep neural net layer, leading to a more optimal estimation. To achieve large temporal receptive fields, multi-scale dilated convolutions are employed to model long-range dependencies among frames. The architecture is straightforward to implement and can be flexibly adopted for real-time applications. Any off-the-shelf 2D pose estimation system, e.g. Mocap libraries, can be easily integrated in an ad-hoc fashion. We both quantitatively and qualitatively evaluate our method on various standard benchmark datasets (e.g. Human3.6M, HumanEva). Our method considerably outperforms all the state-of-the-art algorithms up to 8% error reduction (average mean per joint position error: 34.7) as compared to the best-reported results. Code is available at: (https://github.com/lrxjason/Attention3DHumanPose)","PeriodicalId":6715,"journal":{"name":"2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"23 1","pages":"5063-5072"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77955388","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}
We investigate the emerging task of learning object detectors with sole image-level labels on the web without requiring any other supervision like precise annotations or additional images from well-annotated benchmark datasets. Such a task, termed as fully webly supervised object detection, is extremely challenging, since image-level labels on the web are always noisy, leading to poor performance of the learned detectors. In this work, we propose an end-to-end framework to jointly learn webly supervised detectors and reduce the negative impact of noisy labels. Such noise is heterogeneous, which is further categorized into two types, namely background noise and foreground noise. Regarding the background noise, we propose a residual learning structure incorporated with weakly supervised detection, which decomposes background noise and models clean data. To explicitly learn the residual feature between clean data and noisy labels, we further propose a spatially-sensitive entropy criterion, which exploits the conditional distribution of detection results to estimate the confidence of background categories being noise. Regarding the foreground noise, a bagging-mixup learning is introduced, which suppresses foreground noisy signals from incorrectly labelled images, whilst maintaining the diversity of training data. We evaluate the proposed approach on popular benchmark datasets by training detectors on web images, which are retrieved by the corresponding category tags from photo-sharing sites. Extensive experiments show that our method achieves significant improvements over the state-of-the-art methods.
{"title":"Noise-Aware Fully Webly Supervised Object Detection","authors":"Yunhang Shen, Rongrong Ji, Zhiwei Chen, Xiaopeng Hong, Feng Zheng, Jianzhuang Liu, Mingliang Xu, Q. Tian","doi":"10.1109/cvpr42600.2020.01134","DOIUrl":"https://doi.org/10.1109/cvpr42600.2020.01134","url":null,"abstract":"We investigate the emerging task of learning object detectors with sole image-level labels on the web without requiring any other supervision like precise annotations or additional images from well-annotated benchmark datasets. Such a task, termed as fully webly supervised object detection, is extremely challenging, since image-level labels on the web are always noisy, leading to poor performance of the learned detectors. In this work, we propose an end-to-end framework to jointly learn webly supervised detectors and reduce the negative impact of noisy labels. Such noise is heterogeneous, which is further categorized into two types, namely background noise and foreground noise. Regarding the background noise, we propose a residual learning structure incorporated with weakly supervised detection, which decomposes background noise and models clean data. To explicitly learn the residual feature between clean data and noisy labels, we further propose a spatially-sensitive entropy criterion, which exploits the conditional distribution of detection results to estimate the confidence of background categories being noise. Regarding the foreground noise, a bagging-mixup learning is introduced, which suppresses foreground noisy signals from incorrectly labelled images, whilst maintaining the diversity of training data. We evaluate the proposed approach on popular benchmark datasets by training detectors on web images, which are retrieved by the corresponding category tags from photo-sharing sites. Extensive experiments show that our method achieves significant improvements over the state-of-the-art methods.","PeriodicalId":6715,"journal":{"name":"2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"15 1","pages":"11323-11332"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76256887","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 : 2020-06-01DOI: 10.1109/cvpr42600.2020.00374
Haiyang Mei, Xin Yang, Yang Wang, Yu-An Liu, Shengfeng He, Qiang Zhang, Xiaopeng Wei, Rynson W. H. Lau
Glass is very common in our daily life. Existing computer vision systems neglect it and thus may have severe consequences, e.g., a robot may crash into a glass wall. However, sensing the presence of glass is not straightforward. The key challenge is that arbitrary objects/scenes can appear behind the glass, and the content within the glass region is typically similar to those behind it. In this paper, we propose an important problem of detecting glass from a single RGB image. To address this problem, we construct a large-scale glass detection dataset (GDD) and design a glass detection network, called GDNet, which explores abundant contextual cues for robust glass detection with a novel large-field contextual feature integration (LCFI) module. Extensive experiments demonstrate that the proposed method achieves more superior glass detection results on our GDD test set than state-of-the-art methods fine-tuned for glass detection.
{"title":"Don’t Hit Me! Glass Detection in Real-World Scenes","authors":"Haiyang Mei, Xin Yang, Yang Wang, Yu-An Liu, Shengfeng He, Qiang Zhang, Xiaopeng Wei, Rynson W. H. Lau","doi":"10.1109/cvpr42600.2020.00374","DOIUrl":"https://doi.org/10.1109/cvpr42600.2020.00374","url":null,"abstract":"Glass is very common in our daily life. Existing computer vision systems neglect it and thus may have severe consequences, e.g., a robot may crash into a glass wall. However, sensing the presence of glass is not straightforward. The key challenge is that arbitrary objects/scenes can appear behind the glass, and the content within the glass region is typically similar to those behind it. In this paper, we propose an important problem of detecting glass from a single RGB image. To address this problem, we construct a large-scale glass detection dataset (GDD) and design a glass detection network, called GDNet, which explores abundant contextual cues for robust glass detection with a novel large-field contextual feature integration (LCFI) module. Extensive experiments demonstrate that the proposed method achieves more superior glass detection results on our GDD test set than state-of-the-art methods fine-tuned for glass detection.","PeriodicalId":6715,"journal":{"name":"2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"97 1","pages":"3684-3693"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73285355","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 : 2020-06-01DOI: 10.1109/cvpr42600.2020.00299
Deng-Ping Fan, Zheng Lin, Ge-Peng Ji, Dingwen Zhang, H. Fu, Ming-Ming Cheng
Co-salient object detection (CoSOD) is a newly emerging and rapidly growing branch of salient object detection (SOD), which aims to detect the co-occurring salient objects in multiple images. However, existing CoSOD datasets often have a serious data bias, which assumes that each group of images contains salient objects of similar visual appearances. This bias results in the ideal settings and the effectiveness of the models, trained on existing datasets, may be impaired in real-life situations, where the similarity is usually semantic or conceptual. To tackle this issue, we first collect a new high-quality dataset, named CoSOD3k, which contains 3,316 images divided into 160 groups with multiple level annotations, i.e., category, bounding box, object, and instance levels. CoSOD3k makes a significant leap in terms of diversity, difficulty and scalability, benefiting related vision tasks. Besides, we comprehensively summarize 34 cutting-edge algorithms, benchmarking 19 of them over four existing CoSOD datasets (MSRC, iCoSeg, Image Pair and CoSal2015) and our CoSOD3k with a total of ∼61K images (largest scale), and reporting group-level performance analysis. Finally, we discuss the challenge and future work of CoSOD. Our study would give a strong boost to growth in the CoSOD community. Benchmark toolbox and results are available on our project page.
{"title":"Taking a Deeper Look at Co-Salient Object Detection","authors":"Deng-Ping Fan, Zheng Lin, Ge-Peng Ji, Dingwen Zhang, H. Fu, Ming-Ming Cheng","doi":"10.1109/cvpr42600.2020.00299","DOIUrl":"https://doi.org/10.1109/cvpr42600.2020.00299","url":null,"abstract":"Co-salient object detection (CoSOD) is a newly emerging and rapidly growing branch of salient object detection (SOD), which aims to detect the co-occurring salient objects in multiple images. However, existing CoSOD datasets often have a serious data bias, which assumes that each group of images contains salient objects of similar visual appearances. This bias results in the ideal settings and the effectiveness of the models, trained on existing datasets, may be impaired in real-life situations, where the similarity is usually semantic or conceptual. To tackle this issue, we first collect a new high-quality dataset, named CoSOD3k, which contains 3,316 images divided into 160 groups with multiple level annotations, i.e., category, bounding box, object, and instance levels. CoSOD3k makes a significant leap in terms of diversity, difficulty and scalability, benefiting related vision tasks. Besides, we comprehensively summarize 34 cutting-edge algorithms, benchmarking 19 of them over four existing CoSOD datasets (MSRC, iCoSeg, Image Pair and CoSal2015) and our CoSOD3k with a total of ∼61K images (largest scale), and reporting group-level performance analysis. Finally, we discuss the challenge and future work of CoSOD. Our study would give a strong boost to growth in the CoSOD community. Benchmark toolbox and results are available on our project page.","PeriodicalId":6715,"journal":{"name":"2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"12 1","pages":"2916-2926"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73963587","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 : 2020-06-01DOI: 10.1109/CVPR42600.2020.00307
Yanbei Chen, S. Gong, Loris Bazzani
Image search with text feedback has promising impacts in various real-world applications, such as e-commerce and internet search. Given a reference image and text feedback from user, the goal is to retrieve images that not only resemble the input image, but also change certain aspects in accordance with the given text. This is a challenging task as it requires the synergistic understanding of both image and text. In this work, we tackle this task by a novel Visiolinguistic Attention Learning (VAL) framework. Specifically, we propose a composite transformer that can be seamlessly plugged in a CNN to selectively preserve and transform the visual features conditioned on language semantics. By inserting multiple composite transformers at varying depths, VAL is incentive to encapsulate the multi-granular visiolinguistic information, thus yielding an expressive representation for effective image search. We conduct comprehensive evaluation on three datasets: Fashion200k, Shoes and FashionIQ. Extensive experiments show our model exceeds existing approaches on all datasets, demonstrating consistent superiority in coping with various text feedbacks, including attribute-like and natural language descriptions.
{"title":"Image Search With Text Feedback by Visiolinguistic Attention Learning","authors":"Yanbei Chen, S. Gong, Loris Bazzani","doi":"10.1109/CVPR42600.2020.00307","DOIUrl":"https://doi.org/10.1109/CVPR42600.2020.00307","url":null,"abstract":"Image search with text feedback has promising impacts in various real-world applications, such as e-commerce and internet search. Given a reference image and text feedback from user, the goal is to retrieve images that not only resemble the input image, but also change certain aspects in accordance with the given text. This is a challenging task as it requires the synergistic understanding of both image and text. In this work, we tackle this task by a novel Visiolinguistic Attention Learning (VAL) framework. Specifically, we propose a composite transformer that can be seamlessly plugged in a CNN to selectively preserve and transform the visual features conditioned on language semantics. By inserting multiple composite transformers at varying depths, VAL is incentive to encapsulate the multi-granular visiolinguistic information, thus yielding an expressive representation for effective image search. We conduct comprehensive evaluation on three datasets: Fashion200k, Shoes and FashionIQ. Extensive experiments show our model exceeds existing approaches on all datasets, demonstrating consistent superiority in coping with various text feedbacks, including attribute-like and natural language descriptions.","PeriodicalId":6715,"journal":{"name":"2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"42 1","pages":"2998-3008"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81617299","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 : 2020-06-01DOI: 10.1109/cvpr42600.2020.01379
G. Csurka, Z. Kato, Andor Juhasz, M. Humenberger
Low-rank regions capture geometrically meaningful structures in an image which encompass typical local features such as edges, corners and all kinds of regular, symmetric, often repetitive patterns, that are commonly found in man-made environment. While such patterns are challenging current state-of-the-art feature correspondence methods, the recovered homography of a low-rank texture readily provides 3D structure with respect to a 3D plane, without any prior knowledge of the visual information on that plane. However, the automatic and efficient detection of the broad class of low-rank regions is unsolved. Herein, we propose a novel self-supervised low-rank region detection deep network that predicts a low-rank likelihood map from an image. The evaluation of our method on real-world datasets shows not only that it reliably predicts low-rank regions in the image similarly to our baseline method, but thanks to the data augmentations used in the training phase it generalizes well to difficult cases (e.g. day/night lighting, low contrast, underexposure) where the baseline prediction fails.
{"title":"Estimating Low-Rank Region Likelihood Maps","authors":"G. Csurka, Z. Kato, Andor Juhasz, M. Humenberger","doi":"10.1109/cvpr42600.2020.01379","DOIUrl":"https://doi.org/10.1109/cvpr42600.2020.01379","url":null,"abstract":"Low-rank regions capture geometrically meaningful structures in an image which encompass typical local features such as edges, corners and all kinds of regular, symmetric, often repetitive patterns, that are commonly found in man-made environment. While such patterns are challenging current state-of-the-art feature correspondence methods, the recovered homography of a low-rank texture readily provides 3D structure with respect to a 3D plane, without any prior knowledge of the visual information on that plane. However, the automatic and efficient detection of the broad class of low-rank regions is unsolved. Herein, we propose a novel self-supervised low-rank region detection deep network that predicts a low-rank likelihood map from an image. The evaluation of our method on real-world datasets shows not only that it reliably predicts low-rank regions in the image similarly to our baseline method, but thanks to the data augmentations used in the training phase it generalizes well to difficult cases (e.g. day/night lighting, low contrast, underexposure) where the baseline prediction fails.","PeriodicalId":6715,"journal":{"name":"2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"89 1","pages":"13773-13782"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85143189","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 : 2020-06-01DOI: 10.1109/cvpr42600.2020.01182
Lu Chi, Zehuan Yuan, Yadong Mu, Changhu Wang
Modeling spatial or temporal long-range dependency plays a key role in deep neural networks. Conventional dominant solutions include recurrent operations on sequential data or deeply stacking convolutional layers with small kernel size. Recently, a number of non-local operators (such as self-attention based) have been devised. They are typically generic and can be plugged into many existing network pipelines for globally computing among any two neurons in a feature map. This work proposes a novel non-local operator. It is inspired by the attention mechanism of human visual system, which can quickly attend to important local parts in sight and suppress other less-relevant information. The core of our method is learnable and data-adaptive bilinear attentional transform (BA-Transform), whose merits are three-folds: first, BA-Transform is versatile to model a wide spectrum of local or global attentional operations, such as emphasizing specific local regions. Each BA-Transform is learned in a data-adaptive way; Secondly, to address the discrepancy among features, we further design grouped BA-Transforms, which essentially apply different attentional operations to different groups of feature channels; Thirdly, many existing non-local operators are computation-intensive. The proposed BA-Transform is implemented by simple matrix multiplication and admits better efficacy. For empirical evaluation, we perform comprehensive experiments on two large-scale benchmarks, ImageNet and Kinetics, for image / video classification respectively. The achieved accuracies and various ablation experiments consistently demonstrate significant improvement by large margins.
{"title":"Non-Local Neural Networks With Grouped Bilinear Attentional Transforms","authors":"Lu Chi, Zehuan Yuan, Yadong Mu, Changhu Wang","doi":"10.1109/cvpr42600.2020.01182","DOIUrl":"https://doi.org/10.1109/cvpr42600.2020.01182","url":null,"abstract":"Modeling spatial or temporal long-range dependency plays a key role in deep neural networks. Conventional dominant solutions include recurrent operations on sequential data or deeply stacking convolutional layers with small kernel size. Recently, a number of non-local operators (such as self-attention based) have been devised. They are typically generic and can be plugged into many existing network pipelines for globally computing among any two neurons in a feature map. This work proposes a novel non-local operator. It is inspired by the attention mechanism of human visual system, which can quickly attend to important local parts in sight and suppress other less-relevant information. The core of our method is learnable and data-adaptive bilinear attentional transform (BA-Transform), whose merits are three-folds: first, BA-Transform is versatile to model a wide spectrum of local or global attentional operations, such as emphasizing specific local regions. Each BA-Transform is learned in a data-adaptive way; Secondly, to address the discrepancy among features, we further design grouped BA-Transforms, which essentially apply different attentional operations to different groups of feature channels; Thirdly, many existing non-local operators are computation-intensive. The proposed BA-Transform is implemented by simple matrix multiplication and admits better efficacy. For empirical evaluation, we perform comprehensive experiments on two large-scale benchmarks, ImageNet and Kinetics, for image / video classification respectively. The achieved accuracies and various ablation experiments consistently demonstrate significant improvement by large margins.","PeriodicalId":6715,"journal":{"name":"2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"48 1","pages":"11801-11810"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83969905","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：