Pub Date : 2022-10-05DOI: 10.48550/arXiv.2210.02033
Tianyi Wu, Yusuke Sugano
Human eye contact is a form of non-verbal communication and can have a great influence on social behavior. Since the location and size of the eye contact targets vary across different videos, learning a generic video-independent eye contact detector is still a challenging task. In this work, we address the task of one-way eye contact detection for videos in the wild. Our goal is to build a unified model that can identify when a person is looking at his gaze targets in an arbitrary input video. Considering that this requires time-series relative eye movement information, we propose to formulate the task as a temporal segmentation. Due to the scarcity of labeled training data, we further propose a gaze target discovery method to generate pseudo-labels for unlabeled videos, which allows us to train a generic eye contact segmentation model in an unsupervised way using in-the-wild videos. To evaluate our proposed approach, we manually annotated a test dataset consisting of 52 videos of human conversations. Experimental results show that our eye contact segmentation model outperforms the previous video-dependent eye contact detector and can achieve 71.88% framewise accuracy on our annotated test set. Our code and evaluation dataset are available at https://github.com/ut-vision/Video-Independent-ECS.
{"title":"Learning Video-independent Eye Contact Segmentation from In-the-Wild Videos","authors":"Tianyi Wu, Yusuke Sugano","doi":"10.48550/arXiv.2210.02033","DOIUrl":"https://doi.org/10.48550/arXiv.2210.02033","url":null,"abstract":"Human eye contact is a form of non-verbal communication and can have a great influence on social behavior. Since the location and size of the eye contact targets vary across different videos, learning a generic video-independent eye contact detector is still a challenging task. In this work, we address the task of one-way eye contact detection for videos in the wild. Our goal is to build a unified model that can identify when a person is looking at his gaze targets in an arbitrary input video. Considering that this requires time-series relative eye movement information, we propose to formulate the task as a temporal segmentation. Due to the scarcity of labeled training data, we further propose a gaze target discovery method to generate pseudo-labels for unlabeled videos, which allows us to train a generic eye contact segmentation model in an unsupervised way using in-the-wild videos. To evaluate our proposed approach, we manually annotated a test dataset consisting of 52 videos of human conversations. Experimental results show that our eye contact segmentation model outperforms the previous video-dependent eye contact detector and can achieve 71.88% framewise accuracy on our annotated test set. Our code and evaluation dataset are available at https://github.com/ut-vision/Video-Independent-ECS.","PeriodicalId":87238,"journal":{"name":"Computer vision - ACCV ... : ... Asian Conference on Computer Vision : proceedings. Asian Conference on Computer Vision","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81185808","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 : 2022-10-05DOI: 10.48550/arXiv.2210.02074
Kira Maag, Robin Chan, Svenja Uhlemeyer, K. Kowol, H. Gottschalk
In this work we present two video test data sets for the novel computer vision (CV) task of out of distribution tracking (OOD tracking). Here, OOD objects are understood as objects with a semantic class outside the semantic space of an underlying image segmentation algorithm, or an instance within the semantic space which however looks decisively different from the instances contained in the training data. OOD objects occurring on video sequences should be detected on single frames as early as possible and tracked over their time of appearance as long as possible. During the time of appearance, they should be segmented as precisely as possible. We present the SOS data set containing 20 video sequences of street scenes and more than 1000 labeled frames with up to two OOD objects. We furthermore publish the synthetic CARLA-WildLife data set that consists of 26 video sequences containing up to four OOD objects on a single frame. We propose metrics to measure the success of OOD tracking and develop a baseline algorithm that efficiently tracks the OOD objects. As an application that benefits from OOD tracking, we retrieve OOD sequences from unlabeled videos of street scenes containing OOD objects.
{"title":"Two Video Data Sets for Tracking and Retrieval of Out of Distribution Objects","authors":"Kira Maag, Robin Chan, Svenja Uhlemeyer, K. Kowol, H. Gottschalk","doi":"10.48550/arXiv.2210.02074","DOIUrl":"https://doi.org/10.48550/arXiv.2210.02074","url":null,"abstract":"In this work we present two video test data sets for the novel computer vision (CV) task of out of distribution tracking (OOD tracking). Here, OOD objects are understood as objects with a semantic class outside the semantic space of an underlying image segmentation algorithm, or an instance within the semantic space which however looks decisively different from the instances contained in the training data. OOD objects occurring on video sequences should be detected on single frames as early as possible and tracked over their time of appearance as long as possible. During the time of appearance, they should be segmented as precisely as possible. We present the SOS data set containing 20 video sequences of street scenes and more than 1000 labeled frames with up to two OOD objects. We furthermore publish the synthetic CARLA-WildLife data set that consists of 26 video sequences containing up to four OOD objects on a single frame. We propose metrics to measure the success of OOD tracking and develop a baseline algorithm that efficiently tracks the OOD objects. As an application that benefits from OOD tracking, we retrieve OOD sequences from unlabeled videos of street scenes containing OOD objects.","PeriodicalId":87238,"journal":{"name":"Computer vision - ACCV ... : ... Asian Conference on Computer Vision : proceedings. Asian Conference on Computer Vision","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79743317","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 : 2022-10-05DOI: 10.48550/arXiv.2210.02231
Yue Zhu, David Picard
3D human pose estimation is a challenging task because of the difficulty to acquire ground-truth data outside of controlled environments. A number of further issues have been hindering progress in building a universal and robust model for this task, including domain gaps between different datasets, unseen actions between train and test datasets, various hardware settings and high cost of annotation, etc. In this paper, we propose an algorithm to generate infinite 3D synthetic human poses (Legatus) from a 3D pose distribution based on 10 initial handcrafted 3D poses (Decanus) during the training of a 2D to 3D human pose lifter neural network. Our results show that we can achieve 3D pose estimation performance comparable to methods using real data from specialized datasets but in a zero-shot setup, showing the generalization potential of our framework.
{"title":"Decanus to Legatus: Synthetic training for 2D-3D human pose lifting","authors":"Yue Zhu, David Picard","doi":"10.48550/arXiv.2210.02231","DOIUrl":"https://doi.org/10.48550/arXiv.2210.02231","url":null,"abstract":"3D human pose estimation is a challenging task because of the difficulty to acquire ground-truth data outside of controlled environments. A number of further issues have been hindering progress in building a universal and robust model for this task, including domain gaps between different datasets, unseen actions between train and test datasets, various hardware settings and high cost of annotation, etc. In this paper, we propose an algorithm to generate infinite 3D synthetic human poses (Legatus) from a 3D pose distribution based on 10 initial handcrafted 3D poses (Decanus) during the training of a 2D to 3D human pose lifter neural network. Our results show that we can achieve 3D pose estimation performance comparable to methods using real data from specialized datasets but in a zero-shot setup, showing the generalization potential of our framework.","PeriodicalId":87238,"journal":{"name":"Computer vision - ACCV ... : ... Asian Conference on Computer Vision : proceedings. Asian Conference on Computer Vision","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86327637","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 : 2022-10-05DOI: 10.48550/arXiv.2210.02313
Trong T. Nguyen, Hieu Pham, Phi-Le Nguyen, T. Nguyen, Minh N. Do
Classifying pill categories from real-world images is crucial for various smart healthcare applications. Although existing approaches in image classification might achieve a good performance on fixed pill categories, they fail to handle novel instances of pill categories that are frequently presented to the learning algorithm. To this end, a trivial solution is to train the model with novel classes. However, this may result in a phenomenon known as catastrophic forgetting, in which the system forgets what it learned in previous classes. In this paper, we address this challenge by introducing the class incremental learning (CIL) ability to traditional pill image classification systems. Specifically, we propose a novel incremental multi-stream intermediate fusion framework enabling incorporation of an additional guidance information stream that best matches the domain of the problem into various state-of-the-art CIL methods. From this framework, we consider color-specific information of pill images as a guidance stream and devise an approach, namely"Color Guidance with Multi-stream intermediate fusion"(CG-IMIF) for solving CIL pill image classification task. We conduct comprehensive experiments on real-world incremental pill image classification dataset, namely VAIPE-PCIL, and find that the CG-IMIF consistently outperforms several state-of-the-art methods by a large margin in different task settings. Our code, data, and trained model are available at https://github.com/vinuni-vishc/CG-IMIF.
{"title":"Multi-stream Fusion for Class Incremental Learning in Pill Image Classification","authors":"Trong T. Nguyen, Hieu Pham, Phi-Le Nguyen, T. Nguyen, Minh N. Do","doi":"10.48550/arXiv.2210.02313","DOIUrl":"https://doi.org/10.48550/arXiv.2210.02313","url":null,"abstract":"Classifying pill categories from real-world images is crucial for various smart healthcare applications. Although existing approaches in image classification might achieve a good performance on fixed pill categories, they fail to handle novel instances of pill categories that are frequently presented to the learning algorithm. To this end, a trivial solution is to train the model with novel classes. However, this may result in a phenomenon known as catastrophic forgetting, in which the system forgets what it learned in previous classes. In this paper, we address this challenge by introducing the class incremental learning (CIL) ability to traditional pill image classification systems. Specifically, we propose a novel incremental multi-stream intermediate fusion framework enabling incorporation of an additional guidance information stream that best matches the domain of the problem into various state-of-the-art CIL methods. From this framework, we consider color-specific information of pill images as a guidance stream and devise an approach, namely\"Color Guidance with Multi-stream intermediate fusion\"(CG-IMIF) for solving CIL pill image classification task. We conduct comprehensive experiments on real-world incremental pill image classification dataset, namely VAIPE-PCIL, and find that the CG-IMIF consistently outperforms several state-of-the-art methods by a large margin in different task settings. Our code, data, and trained model are available at https://github.com/vinuni-vishc/CG-IMIF.","PeriodicalId":87238,"journal":{"name":"Computer vision - ACCV ... : ... Asian Conference on Computer Vision : proceedings. Asian Conference on Computer Vision","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73966541","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 : 2022-10-04DOI: 10.48550/arXiv.2210.01485
Yuncheng Jiang, Zixun Zhang, Shixi Qin, Yao Guo, Zhuguo Li, Shuguang Cui
In 3D medical image segmentation, small targets segmentation is crucial for diagnosis but still faces challenges. In this paper, we propose the Axis Projection Attention UNet, named APAUNet, for 3D medical image segmentation, especially for small targets. Considering the large proportion of the background in the 3D feature space, we introduce a projection strategy to project the 3D features into three orthogonal 2D planes to capture the contextual attention from different views. In this way, we can filter out the redundant feature information and mitigate the loss of critical information for small lesions in 3D scans. Then we utilize a dimension hybridization strategy to fuse the 3D features with attention from different axes and merge them by a weighted summation to adaptively learn the importance of different perspectives. Finally, in the APA Decoder, we concatenate both high and low resolution features in the 2D projection process, thereby obtaining more precise multi-scale information, which is vital for small lesion segmentation. Quantitative and qualitative experimental results on two public datasets (BTCV and MSD) demonstrate that our proposed APAUNet outperforms the other methods. Concretely, our APAUNet achieves an average dice score of 87.84 on BTCV, 84.48 on MSD-Liver and 69.13 on MSD-Pancreas, and significantly surpass the previous SOTA methods on small targets.
{"title":"APAUNet: Axis Projection Attention UNet for Small Target in 3D Medical Segmentation","authors":"Yuncheng Jiang, Zixun Zhang, Shixi Qin, Yao Guo, Zhuguo Li, Shuguang Cui","doi":"10.48550/arXiv.2210.01485","DOIUrl":"https://doi.org/10.48550/arXiv.2210.01485","url":null,"abstract":"In 3D medical image segmentation, small targets segmentation is crucial for diagnosis but still faces challenges. In this paper, we propose the Axis Projection Attention UNet, named APAUNet, for 3D medical image segmentation, especially for small targets. Considering the large proportion of the background in the 3D feature space, we introduce a projection strategy to project the 3D features into three orthogonal 2D planes to capture the contextual attention from different views. In this way, we can filter out the redundant feature information and mitigate the loss of critical information for small lesions in 3D scans. Then we utilize a dimension hybridization strategy to fuse the 3D features with attention from different axes and merge them by a weighted summation to adaptively learn the importance of different perspectives. Finally, in the APA Decoder, we concatenate both high and low resolution features in the 2D projection process, thereby obtaining more precise multi-scale information, which is vital for small lesion segmentation. Quantitative and qualitative experimental results on two public datasets (BTCV and MSD) demonstrate that our proposed APAUNet outperforms the other methods. Concretely, our APAUNet achieves an average dice score of 87.84 on BTCV, 84.48 on MSD-Liver and 69.13 on MSD-Pancreas, and significantly surpass the previous SOTA methods on small targets.","PeriodicalId":87238,"journal":{"name":"Computer vision - ACCV ... : ... Asian Conference on Computer Vision : proceedings. Asian Conference on Computer Vision","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80871019","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 : 2022-10-03DOI: 10.48550/arXiv.2210.00757
Tianyu Yan, Zifu Wan, Pingping Zhang
Recently, change detection (CD) of remote sensing images have achieved great progress with the advances of deep learning. However, current methods generally deliver incomplete CD regions and irregular CD boundaries due to the limited representation ability of the extracted visual features. To relieve these issues, in this work we propose a novel learning framework named Fully Transformer Network (FTN) for remote sensing image CD, which improves the feature extraction from a global view and combines multi-level visual features in a pyramid manner. More specifically, the proposed framework first utilizes the advantages of Transformers in long-range dependency modeling. It can help to learn more discriminative global-level features and obtain complete CD regions. Then, we introduce a pyramid structure to aggregate multi-level visual features from Transformers for feature enhancement. The pyramid structure grafted with a Progressive Attention Module (PAM) can improve the feature representation ability with additional interdependencies through channel attentions. Finally, to better train the framework, we utilize the deeply-supervised learning with multiple boundaryaware loss functions. Extensive experiments demonstrate that our proposed method achieves a new state-of-the-art performance on four public CD benchmarks. For model reproduction, the source code is released at https://github.com/AI-Zhpp/FTN.
{"title":"Fully Transformer Network for Change Detection of Remote Sensing Images","authors":"Tianyu Yan, Zifu Wan, Pingping Zhang","doi":"10.48550/arXiv.2210.00757","DOIUrl":"https://doi.org/10.48550/arXiv.2210.00757","url":null,"abstract":"Recently, change detection (CD) of remote sensing images have achieved great progress with the advances of deep learning. However, current methods generally deliver incomplete CD regions and irregular CD boundaries due to the limited representation ability of the extracted visual features. To relieve these issues, in this work we propose a novel learning framework named Fully Transformer Network (FTN) for remote sensing image CD, which improves the feature extraction from a global view and combines multi-level visual features in a pyramid manner. More specifically, the proposed framework first utilizes the advantages of Transformers in long-range dependency modeling. It can help to learn more discriminative global-level features and obtain complete CD regions. Then, we introduce a pyramid structure to aggregate multi-level visual features from Transformers for feature enhancement. The pyramid structure grafted with a Progressive Attention Module (PAM) can improve the feature representation ability with additional interdependencies through channel attentions. Finally, to better train the framework, we utilize the deeply-supervised learning with multiple boundaryaware loss functions. Extensive experiments demonstrate that our proposed method achieves a new state-of-the-art performance on four public CD benchmarks. For model reproduction, the source code is released at https://github.com/AI-Zhpp/FTN.","PeriodicalId":87238,"journal":{"name":"Computer vision - ACCV ... : ... Asian Conference on Computer Vision : proceedings. Asian Conference on Computer Vision","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76803016","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 : 2022-09-29DOI: 10.48550/arXiv.2209.14819
Xingyi Li, Chaoyi Hong, Yiran Wang, Z. Cao, Ke Xian, Guosheng Lin
We study the problem of novel view synthesis of objects from a single image. Existing methods have demonstrated the potential in single-view view synthesis. However, they still fail to recover the fine appearance details, especially in self-occluded areas. This is because a single view only provides limited information. We observe that manmade objects usually exhibit symmetric appearances, which introduce additional prior knowledge. Motivated by this, we investigate the potential performance gains of explicitly embedding symmetry into the scene representation. In this paper, we propose SymmNeRF, a neural radiance field (NeRF) based framework that combines local and global conditioning under the introduction of symmetry priors. In particular, SymmNeRF takes the pixel-aligned image features and the corresponding symmetric features as extra inputs to the NeRF, whose parameters are generated by a hypernetwork. As the parameters are conditioned on the image-encoded latent codes, SymmNeRF is thus scene-independent and can generalize to new scenes. Experiments on synthetic and real-world datasets show that SymmNeRF synthesizes novel views with more details regardless of the pose transformation, and demonstrates good generalization when applied to unseen objects. Code is available at: https://github.com/xingyi-li/SymmNeRF.
{"title":"SymmNeRF: Learning to Explore Symmetry Prior for Single-View View Synthesis","authors":"Xingyi Li, Chaoyi Hong, Yiran Wang, Z. Cao, Ke Xian, Guosheng Lin","doi":"10.48550/arXiv.2209.14819","DOIUrl":"https://doi.org/10.48550/arXiv.2209.14819","url":null,"abstract":"We study the problem of novel view synthesis of objects from a single image. Existing methods have demonstrated the potential in single-view view synthesis. However, they still fail to recover the fine appearance details, especially in self-occluded areas. This is because a single view only provides limited information. We observe that manmade objects usually exhibit symmetric appearances, which introduce additional prior knowledge. Motivated by this, we investigate the potential performance gains of explicitly embedding symmetry into the scene representation. In this paper, we propose SymmNeRF, a neural radiance field (NeRF) based framework that combines local and global conditioning under the introduction of symmetry priors. In particular, SymmNeRF takes the pixel-aligned image features and the corresponding symmetric features as extra inputs to the NeRF, whose parameters are generated by a hypernetwork. As the parameters are conditioned on the image-encoded latent codes, SymmNeRF is thus scene-independent and can generalize to new scenes. Experiments on synthetic and real-world datasets show that SymmNeRF synthesizes novel views with more details regardless of the pose transformation, and demonstrates good generalization when applied to unseen objects. Code is available at: https://github.com/xingyi-li/SymmNeRF.","PeriodicalId":87238,"journal":{"name":"Computer vision - ACCV ... : ... Asian Conference on Computer Vision : proceedings. Asian Conference on Computer Vision","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74844097","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 : 2022-09-28DOI: 10.48550/arXiv.2209.13853
Nasib Ullah, Partha Pratim Mohanta
With the advent of rich visual representations and pre-trained language models, video captioning has seen continuous improvement over time. Despite the performance improvement, video captioning models are prone to hallucination. Hallucination refers to the generation of highly pathological descriptions that are detached from the source material. In video captioning, there are two kinds of hallucination: object and action hallucination. Instead of endeavoring to learn better representations of a video, in this work, we investigate the fundamental sources of the hallucination problem. We identify three main factors: (i) inadequate visual features extracted from pre-trained models, (ii) improper influences of source and target contexts during multi-modal fusion, and (iii) exposure bias in the training strategy. To alleviate these problems, we propose two robust solutions: (a) the introduction of auxiliary heads trained in multi-label settings on top of the extracted visual features and (b) the addition of context gates, which dynamically select the features during fusion. The standard evaluation metrics for video captioning measures similarity with ground truth captions and do not adequately capture object and action relevance. To this end, we propose a new metric, COAHA (caption object and action hallucination assessment), which assesses the degree of hallucination. Our method achieves state-of-the-art performance on the MSR-Video to Text (MSR-VTT) and the Microsoft Research Video Description Corpus (MSVD) datasets, especially by a massive margin in CIDEr score.
{"title":"Thinking Hallucination for Video Captioning","authors":"Nasib Ullah, Partha Pratim Mohanta","doi":"10.48550/arXiv.2209.13853","DOIUrl":"https://doi.org/10.48550/arXiv.2209.13853","url":null,"abstract":"With the advent of rich visual representations and pre-trained language models, video captioning has seen continuous improvement over time. Despite the performance improvement, video captioning models are prone to hallucination. Hallucination refers to the generation of highly pathological descriptions that are detached from the source material. In video captioning, there are two kinds of hallucination: object and action hallucination. Instead of endeavoring to learn better representations of a video, in this work, we investigate the fundamental sources of the hallucination problem. We identify three main factors: (i) inadequate visual features extracted from pre-trained models, (ii) improper influences of source and target contexts during multi-modal fusion, and (iii) exposure bias in the training strategy. To alleviate these problems, we propose two robust solutions: (a) the introduction of auxiliary heads trained in multi-label settings on top of the extracted visual features and (b) the addition of context gates, which dynamically select the features during fusion. The standard evaluation metrics for video captioning measures similarity with ground truth captions and do not adequately capture object and action relevance. To this end, we propose a new metric, COAHA (caption object and action hallucination assessment), which assesses the degree of hallucination. Our method achieves state-of-the-art performance on the MSR-Video to Text (MSR-VTT) and the Microsoft Research Video Description Corpus (MSVD) datasets, especially by a massive margin in CIDEr score.","PeriodicalId":87238,"journal":{"name":"Computer vision - ACCV ... : ... Asian Conference on Computer Vision : proceedings. Asian Conference on Computer Vision","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75727666","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 : 2022-09-28DOI: 10.48550/arXiv.2209.14099
Jiaguo Yu, Huming Qiu, Dubing Chen, Haofeng Zhang
The development of unsupervised hashing is advanced by the recent popular contrastive learning paradigm. However, previous contrastive learning-based works have been hampered by (1) insufficient data similarity mining based on global-only image representations, and (2) the hash code semantic loss caused by the data augmentation. In this paper, we propose a novel method, namely Weighted Contrative Hashing (WCH), to take a step towards solving these two problems. We introduce a novel mutual attention module to alleviate the problem of information asymmetry in network features caused by the missing image structure during contrative augmentation. Furthermore, we explore the fine-grained semantic relations between images, i.e., we divide the images into multiple patches and calculate similarities between patches. The aggregated weighted similarities, which reflect the deep image relations, are distilled to facilitate the hash codes learning with a distillation loss, so as to obtain better retrieval performance. Extensive experiments show that the proposed WCH significantly outperforms existing unsupervised hashing methods on three benchmark datasets.
{"title":"Weighted Contrastive Hashing","authors":"Jiaguo Yu, Huming Qiu, Dubing Chen, Haofeng Zhang","doi":"10.48550/arXiv.2209.14099","DOIUrl":"https://doi.org/10.48550/arXiv.2209.14099","url":null,"abstract":"The development of unsupervised hashing is advanced by the recent popular contrastive learning paradigm. However, previous contrastive learning-based works have been hampered by (1) insufficient data similarity mining based on global-only image representations, and (2) the hash code semantic loss caused by the data augmentation. In this paper, we propose a novel method, namely Weighted Contrative Hashing (WCH), to take a step towards solving these two problems. We introduce a novel mutual attention module to alleviate the problem of information asymmetry in network features caused by the missing image structure during contrative augmentation. Furthermore, we explore the fine-grained semantic relations between images, i.e., we divide the images into multiple patches and calculate similarities between patches. The aggregated weighted similarities, which reflect the deep image relations, are distilled to facilitate the hash codes learning with a distillation loss, so as to obtain better retrieval performance. Extensive experiments show that the proposed WCH significantly outperforms existing unsupervised hashing methods on three benchmark datasets.","PeriodicalId":87238,"journal":{"name":"Computer vision - ACCV ... : ... Asian Conference on Computer Vision : proceedings. Asian Conference on Computer Vision","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80358768","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 : 2022-09-27DOI: 10.48550/arXiv.2209.13378
Xiatao Kang, P. Li, Jiayi Yao, Chengxi Li
Pruning on neural networks before training not only compresses the original models, but also accelerates the network training phase, which has substantial application value. The current work focuses on fine-grained pruning, which uses metrics to calculate weight scores for weight screening, and extends from the initial single-order pruning to iterative pruning. Through these works, we argue that network pruning can be summarized as an expressive force transfer process of weights, where the reserved weights will take on the expressive force from the removed ones for the purpose of maintaining the performance of original networks. In order to achieve optimal expressive force scheduling, we propose a pruning scheme before training called Neural Network Panning which guides expressive force transfer through multi-index and multi-process steps, and designs a kind of panning agent based on reinforcement learning to automate processes. Experimental results show that Panning performs better than various available pruning before training methods.
{"title":"Neural Network Panning: Screening the Optimal Sparse Network Before Training","authors":"Xiatao Kang, P. Li, Jiayi Yao, Chengxi Li","doi":"10.48550/arXiv.2209.13378","DOIUrl":"https://doi.org/10.48550/arXiv.2209.13378","url":null,"abstract":"Pruning on neural networks before training not only compresses the original models, but also accelerates the network training phase, which has substantial application value. The current work focuses on fine-grained pruning, which uses metrics to calculate weight scores for weight screening, and extends from the initial single-order pruning to iterative pruning. Through these works, we argue that network pruning can be summarized as an expressive force transfer process of weights, where the reserved weights will take on the expressive force from the removed ones for the purpose of maintaining the performance of original networks. In order to achieve optimal expressive force scheduling, we propose a pruning scheme before training called Neural Network Panning which guides expressive force transfer through multi-index and multi-process steps, and designs a kind of panning agent based on reinforcement learning to automate processes. Experimental results show that Panning performs better than various available pruning before training methods.","PeriodicalId":87238,"journal":{"name":"Computer vision - ACCV ... : ... Asian Conference on Computer Vision : proceedings. Asian Conference on Computer Vision","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77412133","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}