Video object detection is a fundamental tool for many applications. Since direct application of image-based object detection cannot leverage the rich temporal information inherent in video data, we advocate to the detection of long-range video object pattern. While the Long Short-Term Memory (LSTM) has been the de facto choice for such detection, currently LSTM cannot fundamentally model object association between consecutive frames. In this paper, we propose the association LSTM to address this fundamental association problem. Association LSTM not only regresses and classifiy directly on object locations and categories but also associates features to represent each output object. By minimizing the matching error between these features, we learn how to associate objects in two consecutive frames. Additionally, our method works in an online manner, which is important for most video tasks. Compared to the traditional video object detection methods, our approach outperforms them on standard video datasets.
{"title":"Online Video Object Detection Using Association LSTM","authors":"Yongyi Lu, Cewu Lu, Chi-Keung Tang","doi":"10.1109/ICCV.2017.257","DOIUrl":"https://doi.org/10.1109/ICCV.2017.257","url":null,"abstract":"Video object detection is a fundamental tool for many applications. Since direct application of image-based object detection cannot leverage the rich temporal information inherent in video data, we advocate to the detection of long-range video object pattern. While the Long Short-Term Memory (LSTM) has been the de facto choice for such detection, currently LSTM cannot fundamentally model object association between consecutive frames. In this paper, we propose the association LSTM to address this fundamental association problem. Association LSTM not only regresses and classifiy directly on object locations and categories but also associates features to represent each output object. By minimizing the matching error between these features, we learn how to associate objects in two consecutive frames. Additionally, our method works in an online manner, which is important for most video tasks. Compared to the traditional video object detection methods, our approach outperforms them on standard video datasets.","PeriodicalId":6559,"journal":{"name":"2017 IEEE International Conference on Computer Vision (ICCV)","volume":"107 1","pages":"2363-2371"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73700276","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}
Research into computational jigsaw puzzle solving, an emerging theoretical problem with numerous applications, has focused in recent years on puzzles that constitute square pieces only. In this paper we wish to extend the scientific scope of appearance-based puzzle solving and consider ’’brick wall” jigsaw puzzles – rectangular pieces who may have different sizes, and could be placed next to each other at arbitrary offset along their abutting edge – a more explicit configuration with properties of real world puzzles. We present the new challenges that arise in brick wall puzzles and address them in two stages. First we concentrate on the reconstruction of the puzzle (with or without missing pieces) assuming an oracle for offset assignments. We show that despite the increased complexity of the problem, under these conditions performance can be made comparable to the state-of-the-art in solving the simpler square piece puzzles, and thereby argue that solving brick wall puzzles may be reduced to finding the correct offset between two neighboring pieces. We then move on to focus on implementing the oracle computationally using a mixture of dissimilarity metrics and correlation matching. We show results on various brick wall puzzles and discuss how our work may start a new research path for the puzzle solving community.
{"title":"From Square Pieces to Brick Walls: The Next Challenge in Solving Jigsaw Puzzles","authors":"Shir Gur, O. Ben-Shahar","doi":"10.1109/ICCV.2017.434","DOIUrl":"https://doi.org/10.1109/ICCV.2017.434","url":null,"abstract":"Research into computational jigsaw puzzle solving, an emerging theoretical problem with numerous applications, has focused in recent years on puzzles that constitute square pieces only. In this paper we wish to extend the scientific scope of appearance-based puzzle solving and consider ’’brick wall” jigsaw puzzles – rectangular pieces who may have different sizes, and could be placed next to each other at arbitrary offset along their abutting edge – a more explicit configuration with properties of real world puzzles. We present the new challenges that arise in brick wall puzzles and address them in two stages. First we concentrate on the reconstruction of the puzzle (with or without missing pieces) assuming an oracle for offset assignments. We show that despite the increased complexity of the problem, under these conditions performance can be made comparable to the state-of-the-art in solving the simpler square piece puzzles, and thereby argue that solving brick wall puzzles may be reduced to finding the correct offset between two neighboring pieces. We then move on to focus on implementing the oracle computationally using a mixture of dissimilarity metrics and correlation matching. We show results on various brick wall puzzles and discuss how our work may start a new research path for the puzzle solving community.","PeriodicalId":6559,"journal":{"name":"2017 IEEE International Conference on Computer Vision (ICCV)","volume":"5 1","pages":"4049-4057"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74256485","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}
In this work we propose a novel framework named Dual-Net aiming at learning more accurate representation for image recognition. Here two parallel neural networks are coordinated to learn complementary features and thus a wider network is constructed. Specifically, we logically divide an end-to-end deep convolutional neural network into two functional parts, i.e., feature extractor and image classifier. The extractors of two subnetworks are placed side by side, which exactly form the feature extractor of DualNet. Then the two-stream features are aggregated to the final classifier for overall classification, while two auxiliary classifiers are appended behind the feature extractor of each subnetwork to make the separately learned features discriminative alone. The complementary constraint is imposed by weighting the three classifiers, which is indeed the key of DualNet. The corresponding training strategy is also proposed, consisting of iterative training and joint finetuning, to make the two subnetworks cooperate well with each other. Finally, DualNet based on the well-known CaffeNet, VGGNet, NIN and ResNet are thoroughly investigated and experimentally evaluated on multiple datasets including CIFAR-100, Stanford Dogs and UEC FOOD-100. The results demonstrate that DualNet can really help learn more accurate image representation, and thus result in higher accuracy for recognition. In particular, the performance on CIFAR-100 is state-of-the-art compared to the recent works.
{"title":"DualNet: Learn Complementary Features for Image Recognition","authors":"Saihui Hou, X. Liu, Zilei Wang","doi":"10.1109/ICCV.2017.62","DOIUrl":"https://doi.org/10.1109/ICCV.2017.62","url":null,"abstract":"In this work we propose a novel framework named Dual-Net aiming at learning more accurate representation for image recognition. Here two parallel neural networks are coordinated to learn complementary features and thus a wider network is constructed. Specifically, we logically divide an end-to-end deep convolutional neural network into two functional parts, i.e., feature extractor and image classifier. The extractors of two subnetworks are placed side by side, which exactly form the feature extractor of DualNet. Then the two-stream features are aggregated to the final classifier for overall classification, while two auxiliary classifiers are appended behind the feature extractor of each subnetwork to make the separately learned features discriminative alone. The complementary constraint is imposed by weighting the three classifiers, which is indeed the key of DualNet. The corresponding training strategy is also proposed, consisting of iterative training and joint finetuning, to make the two subnetworks cooperate well with each other. Finally, DualNet based on the well-known CaffeNet, VGGNet, NIN and ResNet are thoroughly investigated and experimentally evaluated on multiple datasets including CIFAR-100, Stanford Dogs and UEC FOOD-100. The results demonstrate that DualNet can really help learn more accurate image representation, and thus result in higher accuracy for recognition. In particular, the performance on CIFAR-100 is state-of-the-art compared to the recent works.","PeriodicalId":6559,"journal":{"name":"2017 IEEE International Conference on Computer Vision (ICCV)","volume":"123 1","pages":"502-510"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74797457","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}
Most modern face super-resolution methods resort to convolutional neural networks (CNN) to infer highresolution (HR) face images. When dealing with very low resolution (LR) images, the performance of these CNN based methods greatly degrades. Meanwhile, these methods tend to produce over-smoothed outputs and miss some textural details. To address these challenges, this paper presents a wavelet-based CNN approach that can ultra-resolve a very low resolution face image of 16 × 16 or smaller pixelsize to its larger version of multiple scaling factors (2×, 4×, 8× and even 16×) in a unified framework. Different from conventional CNN methods directly inferring HR images, our approach firstly learns to predict the LR’s corresponding series of HR’s wavelet coefficients before reconstructing HR images from them. To capture both global topology information and local texture details of human faces, we present a flexible and extensible convolutional neural network with three types of loss: wavelet prediction loss, texture loss and full-image loss. Extensive experiments demonstrate that the proposed approach achieves more appealing results both quantitatively and qualitatively than state-ofthe- art super-resolution methods.
{"title":"Wavelet-SRNet: A Wavelet-Based CNN for Multi-scale Face Super Resolution","authors":"Huaibo Huang, R. He, Zhenan Sun, T. Tan","doi":"10.1109/ICCV.2017.187","DOIUrl":"https://doi.org/10.1109/ICCV.2017.187","url":null,"abstract":"Most modern face super-resolution methods resort to convolutional neural networks (CNN) to infer highresolution (HR) face images. When dealing with very low resolution (LR) images, the performance of these CNN based methods greatly degrades. Meanwhile, these methods tend to produce over-smoothed outputs and miss some textural details. To address these challenges, this paper presents a wavelet-based CNN approach that can ultra-resolve a very low resolution face image of 16 × 16 or smaller pixelsize to its larger version of multiple scaling factors (2×, 4×, 8× and even 16×) in a unified framework. Different from conventional CNN methods directly inferring HR images, our approach firstly learns to predict the LR’s corresponding series of HR’s wavelet coefficients before reconstructing HR images from them. To capture both global topology information and local texture details of human faces, we present a flexible and extensible convolutional neural network with three types of loss: wavelet prediction loss, texture loss and full-image loss. Extensive experiments demonstrate that the proposed approach achieves more appealing results both quantitatively and qualitatively than state-ofthe- art super-resolution methods.","PeriodicalId":6559,"journal":{"name":"2017 IEEE International Conference on Computer Vision (ICCV)","volume":"55 1","pages":"1698-1706"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84490771","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}
Person re-identification is a challenge in video-based surveillance where the goal is to identify the same person in different camera views. In recent years, many algorithms have been proposed that approach this problem by designing suitable feature representations for images of persons or by training appropriate distance metrics that learn to distinguish between images of different persons. Aggregating the results from multiple algorithms for person re-identification is a relatively less-explored area of research. In this paper, we formulate an algorithm that maps the ranking process in a person re-identification algorithm to a problem in graph theory. We then extend this formulation to allow for the use of results from multiple algorithms to make a consensus-based decision for the person re-identification problem. The algorithm is unsupervised and takes into account only the matching scores generated by multiple algorithms for creating a consensus of results. Further, we show how the graph theoretic problem can be solved by a two-step process. First, we obtain a rough estimate of the solution using a greedy algorithm. Then, we extend the construction of the proposed graph so that the problem can be efficiently solved by means of Ant Colony Optimization, a heuristic path-searching algorithm for complex graphs. While we present the algorithm in the context of person reidentification, it can potentially be applied to the general problem of ranking items based on a consensus of multiple sets of scores or metric values.
{"title":"SHaPE: A Novel Graph Theoretic Algorithm for Making Consensus-Based Decisions in Person Re-identification Systems","authors":"Arko Barman, S. Shah","doi":"10.1109/ICCV.2017.127","DOIUrl":"https://doi.org/10.1109/ICCV.2017.127","url":null,"abstract":"Person re-identification is a challenge in video-based surveillance where the goal is to identify the same person in different camera views. In recent years, many algorithms have been proposed that approach this problem by designing suitable feature representations for images of persons or by training appropriate distance metrics that learn to distinguish between images of different persons. Aggregating the results from multiple algorithms for person re-identification is a relatively less-explored area of research. In this paper, we formulate an algorithm that maps the ranking process in a person re-identification algorithm to a problem in graph theory. We then extend this formulation to allow for the use of results from multiple algorithms to make a consensus-based decision for the person re-identification problem. The algorithm is unsupervised and takes into account only the matching scores generated by multiple algorithms for creating a consensus of results. Further, we show how the graph theoretic problem can be solved by a two-step process. First, we obtain a rough estimate of the solution using a greedy algorithm. Then, we extend the construction of the proposed graph so that the problem can be efficiently solved by means of Ant Colony Optimization, a heuristic path-searching algorithm for complex graphs. While we present the algorithm in the context of person reidentification, it can potentially be applied to the general problem of ranking items based on a consensus of multiple sets of scores or metric values.","PeriodicalId":6559,"journal":{"name":"2017 IEEE International Conference on Computer Vision (ICCV)","volume":"83 1","pages":"1124-1133"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77220775","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 intensive annotation cost and the rich but unlabeled data contained in videos motivate us to propose an unsupervised video-based person re-identification (re-ID) method. We start from two assumptions: 1) different video tracklets typically contain different persons, given that the tracklets are taken at distinct places or with long intervals; 2) within each tracklet, the frames are mostly of the same person. Based on these assumptions, this paper propose a stepwise metric promotion approach to estimate the identities of training tracklets, which iterates between cross-camera tracklet association and feature learning. Specifically, We use each training tracklet as a query, and perform retrieval in the cross-camera training set. Our method is built on reciprocal nearest neighbor search and can eliminate the hard negative label matches, i.e., the cross-camera nearest neighbors of the false matches in the initial rank list. The tracklet that passes the reciprocal nearest neighbor check is considered to have the same ID with the query. Experimental results on the PRID 2011, ILIDS-VID, and MARS datasets show that the proposed method achieves very competitive re-ID accuracy compared with its supervised counterparts.
{"title":"Stepwise Metric Promotion for Unsupervised Video Person Re-identification","authors":"Zimo Liu, D. Wang, Huchuan Lu","doi":"10.1109/ICCV.2017.266","DOIUrl":"https://doi.org/10.1109/ICCV.2017.266","url":null,"abstract":"The intensive annotation cost and the rich but unlabeled data contained in videos motivate us to propose an unsupervised video-based person re-identification (re-ID) method. We start from two assumptions: 1) different video tracklets typically contain different persons, given that the tracklets are taken at distinct places or with long intervals; 2) within each tracklet, the frames are mostly of the same person. Based on these assumptions, this paper propose a stepwise metric promotion approach to estimate the identities of training tracklets, which iterates between cross-camera tracklet association and feature learning. Specifically, We use each training tracklet as a query, and perform retrieval in the cross-camera training set. Our method is built on reciprocal nearest neighbor search and can eliminate the hard negative label matches, i.e., the cross-camera nearest neighbors of the false matches in the initial rank list. The tracklet that passes the reciprocal nearest neighbor check is considered to have the same ID with the query. Experimental results on the PRID 2011, ILIDS-VID, and MARS datasets show that the proposed method achieves very competitive re-ID accuracy compared with its supervised counterparts.","PeriodicalId":6559,"journal":{"name":"2017 IEEE International Conference on Computer Vision (ICCV)","volume":"145 1","pages":"2448-2457"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78359174","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}
Di Lin, Guangyong Chen, D. Cohen-Or, P. Heng, Hui Huang
Fully convolutional network (FCN) has been successfully applied in semantic segmentation of scenes represented with RGB images. Images augmented with depth channel provide more understanding of the geometric information of the scene in the image. The question is how to best exploit this additional information to improve the segmentation performance.,,In this paper, we present a neural network with multiple branches for segmenting RGB-D images. Our approach is to use the available depth to split the image into layers with common visual characteristic of objects/scenes, or common “scene-resolution”. We introduce context-aware receptive field (CaRF) which provides a better control on the relevant contextual information of the learned features. Equipped with CaRF, each branch of the network semantically segments relevant similar scene-resolution, leading to a more focused domain which is easier to learn. Furthermore, our network is cascaded with features from one branch augmenting the features of adjacent branch. We show that such cascading of features enriches the contextual information of each branch and enhances the overall performance. The accuracy that our network achieves outperforms the stateof-the-art methods on two public datasets.
{"title":"Cascaded Feature Network for Semantic Segmentation of RGB-D Images","authors":"Di Lin, Guangyong Chen, D. Cohen-Or, P. Heng, Hui Huang","doi":"10.1109/ICCV.2017.147","DOIUrl":"https://doi.org/10.1109/ICCV.2017.147","url":null,"abstract":"Fully convolutional network (FCN) has been successfully applied in semantic segmentation of scenes represented with RGB images. Images augmented with depth channel provide more understanding of the geometric information of the scene in the image. The question is how to best exploit this additional information to improve the segmentation performance.,,In this paper, we present a neural network with multiple branches for segmenting RGB-D images. Our approach is to use the available depth to split the image into layers with common visual characteristic of objects/scenes, or common “scene-resolution”. We introduce context-aware receptive field (CaRF) which provides a better control on the relevant contextual information of the learned features. Equipped with CaRF, each branch of the network semantically segments relevant similar scene-resolution, leading to a more focused domain which is easier to learn. Furthermore, our network is cascaded with features from one branch augmenting the features of adjacent branch. We show that such cascading of features enriches the contextual information of each branch and enhances the overall performance. The accuracy that our network achieves outperforms the stateof-the-art methods on two public datasets.","PeriodicalId":6559,"journal":{"name":"2017 IEEE International Conference on Computer Vision (ICCV)","volume":"2014 1","pages":"1320-1328"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73294305","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}
Jiandong Tian, Zak Murez, Tong Cui, Zhen Zhang, D. Kriegman, R. Ramamoorthi
Traditional imaging methods and computer vision algorithms are often ineffective when images are acquired in scattering media, such as underwater, fog, and biological tissue. Here, we explore the use of light field imaging and algorithms for image restoration and depth estimation that address the image degradation from the medium. Towards this end, we make the following three contributions. First, we present a new single image restoration algorithm which removes backscatter and attenuation from images better than existing methods do, and apply it to each view in the light field. Second, we combine a novel transmission based depth cue with existing correspondence and defocus cues to improve light field depth estimation. In densely scattering media, our transmission depth cue is critical for depth estimation since the images have low signal to noise ratios which significantly degrades the performance of the correspondence and defocus cues. Finally, we propose shearing and refocusing multiple views of the light field to recover a single image of higher quality than what is possible from a single view. We demonstrate the benefits of our method through extensive experimental results in a water tank.
{"title":"Depth and Image Restoration from Light Field in a Scattering Medium","authors":"Jiandong Tian, Zak Murez, Tong Cui, Zhen Zhang, D. Kriegman, R. Ramamoorthi","doi":"10.1109/ICCV.2017.263","DOIUrl":"https://doi.org/10.1109/ICCV.2017.263","url":null,"abstract":"Traditional imaging methods and computer vision algorithms are often ineffective when images are acquired in scattering media, such as underwater, fog, and biological tissue. Here, we explore the use of light field imaging and algorithms for image restoration and depth estimation that address the image degradation from the medium. Towards this end, we make the following three contributions. First, we present a new single image restoration algorithm which removes backscatter and attenuation from images better than existing methods do, and apply it to each view in the light field. Second, we combine a novel transmission based depth cue with existing correspondence and defocus cues to improve light field depth estimation. In densely scattering media, our transmission depth cue is critical for depth estimation since the images have low signal to noise ratios which significantly degrades the performance of the correspondence and defocus cues. Finally, we propose shearing and refocusing multiple views of the light field to recover a single image of higher quality than what is possible from a single view. We demonstrate the benefits of our method through extensive experimental results in a water tank.","PeriodicalId":6559,"journal":{"name":"2017 IEEE International Conference on Computer Vision (ICCV)","volume":"1 1","pages":"2420-2429"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80435744","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}
Adrià Recasens, Carl Vondrick, A. Khosla, A. Torralba
Following the gaze of people inside videos is an important signal for understanding people and their actions. In this paper, we present an approach for following gaze in video by predicting where a person (in the video) is looking even when the object is in a different frame. We collect VideoGaze, a new dataset which we use as a benchmark to both train and evaluate models. Given one frame with a person in it, our model estimates a density for gaze location in every frame and the probability that the person is looking in that particular frame. A key aspect of our approach is an end-to-end model that jointly estimates: saliency, gaze pose, and geometric relationships between views while only using gaze as supervision. Visualizations suggest that the model learns to internally solve these intermediate tasks automatically without additional supervision. Experiments show that our approach follows gaze in video better than existing approaches, enabling a richer understanding of human activities in video.
{"title":"Following Gaze in Video","authors":"Adrià Recasens, Carl Vondrick, A. Khosla, A. Torralba","doi":"10.1109/ICCV.2017.160","DOIUrl":"https://doi.org/10.1109/ICCV.2017.160","url":null,"abstract":"Following the gaze of people inside videos is an important signal for understanding people and their actions. In this paper, we present an approach for following gaze in video by predicting where a person (in the video) is looking even when the object is in a different frame. We collect VideoGaze, a new dataset which we use as a benchmark to both train and evaluate models. Given one frame with a person in it, our model estimates a density for gaze location in every frame and the probability that the person is looking in that particular frame. A key aspect of our approach is an end-to-end model that jointly estimates: saliency, gaze pose, and geometric relationships between views while only using gaze as supervision. Visualizations suggest that the model learns to internally solve these intermediate tasks automatically without additional supervision. Experiments show that our approach follows gaze in video better than existing approaches, enabling a richer understanding of human activities in video.","PeriodicalId":6559,"journal":{"name":"2017 IEEE International Conference on Computer Vision (ICCV)","volume":"42 1","pages":"1444-1452"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80450246","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}
A temporal superpixel algorithm based on proximity-weighted patch matching (TS-PPM) is proposed in this work. We develop the proximity-weighted patch matching (PPM), which estimates the motion vector of a superpixel robustly, by considering the patch matching distances of neighboring superpixels as well as the target superpixel. In each frame, we initialize superpixels by transferring the superpixel labels of the previous frame using PPM motion vectors. Then, we update the superpixel labels of boundary pixels, based on a cost function, composed of color, spatial, contour, and temporal consistency terms. Finally, we execute superpixel splitting, merging, and relabeling to regularize superpixel sizes and reduce incorrect labels. Experiments show that the proposed algorithm outperforms the state-of-the-art conventional algorithms significantly.
{"title":"Temporal Superpixels Based on Proximity-Weighted Patch Matching","authors":"Se-Ho Lee, Won-Dong Jang, Chang-Su Kim","doi":"10.1109/ICCV.2017.390","DOIUrl":"https://doi.org/10.1109/ICCV.2017.390","url":null,"abstract":"A temporal superpixel algorithm based on proximity-weighted patch matching (TS-PPM) is proposed in this work. We develop the proximity-weighted patch matching (PPM), which estimates the motion vector of a superpixel robustly, by considering the patch matching distances of neighboring superpixels as well as the target superpixel. In each frame, we initialize superpixels by transferring the superpixel labels of the previous frame using PPM motion vectors. Then, we update the superpixel labels of boundary pixels, based on a cost function, composed of color, spatial, contour, and temporal consistency terms. Finally, we execute superpixel splitting, merging, and relabeling to regularize superpixel sizes and reduce incorrect labels. Experiments show that the proposed algorithm outperforms the state-of-the-art conventional algorithms significantly.","PeriodicalId":6559,"journal":{"name":"2017 IEEE International Conference on Computer Vision (ICCV)","volume":"31 1","pages":"3630-3638"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82583263","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：