Yu Zhang, Dongqing Zou, Jimmy S. J. Ren, Zhe Jiang, Xiaohao Chen
This paper addresses stereoscopic view synthesis from a single image. Various recent works solve this task by reorganizing pixels from the input view to reconstruct the target one in a stereo setup. However, purely depending on such photometric-based reconstruction process, the network may produce structurally inconsistent results. Regarding this issue, this work proposes Multi-Scale Adversarial Correlation Matching (MS-ACM), a novel learning framework for structure-aware view synthesis. The proposed framework does not assume any costly supervision signal of scene structures such as depth. Instead, it models structures as self-correlation coefficients extracted from multi-scale feature maps in transformed spaces. In training, the feature space attempts to push the correlation distances between the synthesized and target images far apart, thus amplifying inconsistent structures. At the same time, the view synthesis network minimizes such correlation distances by fixing mistakes it makes. With such adversarial training, structural errors of different scales and levels are iteratively discovered and reduced, preserving both global layouts and fine-grained details. Extensive experiments on the KITTI benchmark show that MS-ACM improves both visual quality and the metrics over existing methods when plugged into recent view synthesis architectures.
{"title":"Structure-Preserving Stereoscopic View Synthesis With Multi-Scale Adversarial Correlation Matching","authors":"Yu Zhang, Dongqing Zou, Jimmy S. J. Ren, Zhe Jiang, Xiaohao Chen","doi":"10.1109/CVPR.2019.00601","DOIUrl":"https://doi.org/10.1109/CVPR.2019.00601","url":null,"abstract":"This paper addresses stereoscopic view synthesis from a single image. Various recent works solve this task by reorganizing pixels from the input view to reconstruct the target one in a stereo setup. However, purely depending on such photometric-based reconstruction process, the network may produce structurally inconsistent results. Regarding this issue, this work proposes Multi-Scale Adversarial Correlation Matching (MS-ACM), a novel learning framework for structure-aware view synthesis. The proposed framework does not assume any costly supervision signal of scene structures such as depth. Instead, it models structures as self-correlation coefficients extracted from multi-scale feature maps in transformed spaces. In training, the feature space attempts to push the correlation distances between the synthesized and target images far apart, thus amplifying inconsistent structures. At the same time, the view synthesis network minimizes such correlation distances by fixing mistakes it makes. With such adversarial training, structural errors of different scales and levels are iteratively discovered and reduced, preserving both global layouts and fine-grained details. Extensive experiments on the KITTI benchmark show that MS-ACM improves both visual quality and the metrics over existing methods when plugged into recent view synthesis architectures.","PeriodicalId":6711,"journal":{"name":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"31 1","pages":"5853-5862"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74514968","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}
Conventionally, deep neural networks are trained offline, relying on a large dataset prepared in advance. This paradigm is often challenged in real-world applications, e.g. online services that involve continuous streams of incoming data. Recently, incremental learning receives increasing attention, and is considered as a promising solution to the practical challenges mentioned above. However, it has been observed that incremental learning is subject to a fundamental difficulty -- catastrophic forgetting, namely adapting a model to new data often results in severe performance degradation on previous tasks or classes. Our study reveals that the imbalance between previous and new data is a crucial cause to this problem. In this work, we develop a new framework for incrementally learning a unified classifier, e.g. a classifier that treats both old and new classes uniformly. Specifically, we incorporate three components, cosine normalization, less-forget constraint, and inter-class separation, to mitigate the adverse effects of the imbalance. Experiments show that the proposed method can effectively rebalance the training process, thus obtaining superior performance compared to the existing methods. On CIFAR-100 and ImageNet, our method can reduce the classification errors by more than 6% and 13% respectively, under the incremental setting of 10 phases.
{"title":"Learning a Unified Classifier Incrementally via Rebalancing","authors":"Saihui Hou, Xinyu Pan, Chen Change Loy, Zilei Wang, Dahua Lin","doi":"10.1109/CVPR.2019.00092","DOIUrl":"https://doi.org/10.1109/CVPR.2019.00092","url":null,"abstract":"Conventionally, deep neural networks are trained offline, relying on a large dataset prepared in advance. This paradigm is often challenged in real-world applications, e.g. online services that involve continuous streams of incoming data. Recently, incremental learning receives increasing attention, and is considered as a promising solution to the practical challenges mentioned above. However, it has been observed that incremental learning is subject to a fundamental difficulty -- catastrophic forgetting, namely adapting a model to new data often results in severe performance degradation on previous tasks or classes. Our study reveals that the imbalance between previous and new data is a crucial cause to this problem. In this work, we develop a new framework for incrementally learning a unified classifier, e.g. a classifier that treats both old and new classes uniformly. Specifically, we incorporate three components, cosine normalization, less-forget constraint, and inter-class separation, to mitigate the adverse effects of the imbalance. Experiments show that the proposed method can effectively rebalance the training process, thus obtaining superior performance compared to the existing methods. On CIFAR-100 and ImageNet, our method can reduce the classification errors by more than 6% and 13% respectively, under the incremental setting of 10 phases.","PeriodicalId":6711,"journal":{"name":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"51 1","pages":"831-839"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74530982","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}
Recognizing every person's action in a crowded and cluttered environment is a challenging task in computer vision. We propose to tackle this challenging problem using a holistic 4D ``scan'' of a cluttered scene to include every detail about the people and environment. This leads to a new problem, i.e., recognizing multiple people's actions in the cluttered 4D representation. At the first step, we propose a new method to track people in 4D, which can reliably detect and follow each person in real time. Then, we build a new deep neural network, the Action4DNet, to recognize the action of each tracked person. Such a model gives reliable and accurate results in the real-world settings. We also design an adaptive 3D convolution layer and a novel discriminative temporal feature learning objective to further improve the performance of our model. Our method is invariant to camera view angles, resistant to clutter and able to handle crowd. The experimental results show that the proposed method is fast, reliable and accurate. Our method paves the way to action recognition in the real-world applications and is ready to be deployed to enable smart homes, smart factories and smart stores.
{"title":"Action4D: Online Action Recognition in the Crowd and Clutter","authors":"Quanzeng You, Hao Jiang","doi":"10.1109/CVPR.2019.01213","DOIUrl":"https://doi.org/10.1109/CVPR.2019.01213","url":null,"abstract":"Recognizing every person's action in a crowded and cluttered environment is a challenging task in computer vision. We propose to tackle this challenging problem using a holistic 4D ``scan'' of a cluttered scene to include every detail about the people and environment. This leads to a new problem, i.e., recognizing multiple people's actions in the cluttered 4D representation. At the first step, we propose a new method to track people in 4D, which can reliably detect and follow each person in real time. Then, we build a new deep neural network, the Action4DNet, to recognize the action of each tracked person. Such a model gives reliable and accurate results in the real-world settings. We also design an adaptive 3D convolution layer and a novel discriminative temporal feature learning objective to further improve the performance of our model. Our method is invariant to camera view angles, resistant to clutter and able to handle crowd. The experimental results show that the proposed method is fast, reliable and accurate. Our method paves the way to action recognition in the real-world applications and is ready to be deployed to enable smart homes, smart factories and smart stores.","PeriodicalId":6711,"journal":{"name":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"26 1","pages":"11849-11858"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72941666","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}
Bo Jiang, Ziyan Zhang, Doudou Lin, Jin Tang, B. Luo
Graph Convolutional Neural Networks (graph CNNs) have been widely used for graph data representation and semi-supervised learning tasks. However, existing graph CNNs generally use a fixed graph which may not be optimal for semi-supervised learning tasks. In this paper, we propose a novel Graph Learning-Convolutional Network (GLCN) for graph data representation and semi-supervised learning. The aim of GLCN is to learn an optimal graph structure that best serves graph CNNs for semi-supervised learning by integrating both graph learning and graph convolution in a unified network architecture. The main advantage is that in GLCN both given labels and the estimated labels are incorporated and thus can provide useful ‘weakly’ supervised information to refine (or learn) the graph construction and also to facilitate the graph convolution operation for unknown label estimation. Experimental results on seven benchmarks demonstrate that GLCN significantly outperforms the state-of-the-art traditional fixed structure based graph CNNs.
{"title":"Semi-Supervised Learning With Graph Learning-Convolutional Networks","authors":"Bo Jiang, Ziyan Zhang, Doudou Lin, Jin Tang, B. Luo","doi":"10.1109/CVPR.2019.01157","DOIUrl":"https://doi.org/10.1109/CVPR.2019.01157","url":null,"abstract":"Graph Convolutional Neural Networks (graph CNNs) have been widely used for graph data representation and semi-supervised learning tasks. However, existing graph CNNs generally use a fixed graph which may not be optimal for semi-supervised learning tasks. In this paper, we propose a novel Graph Learning-Convolutional Network (GLCN) for graph data representation and semi-supervised learning. The aim of GLCN is to learn an optimal graph structure that best serves graph CNNs for semi-supervised learning by integrating both graph learning and graph convolution in a unified network architecture. The main advantage is that in GLCN both given labels and the estimated labels are incorporated and thus can provide useful ‘weakly’ supervised information to refine (or learn) the graph construction and also to facilitate the graph convolution operation for unknown label estimation. Experimental results on seven benchmarks demonstrate that GLCN significantly outperforms the state-of-the-art traditional fixed structure based graph CNNs.","PeriodicalId":6711,"journal":{"name":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"4 1","pages":"11305-11312"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75421011","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}
N. Neverova, James Thewlis, R. Güler, Iasonas Kokkinos, A. Vedaldi
DensePose supersedes traditional landmark detectors by densely mapping image pixels to body surface coordinates. This power, however, comes at a greatly increased annotation cost, as supervising the model requires to manually label hundreds of points per pose instance. In this work, we thus seek methods to significantly slim down the DensePose annotations, proposing more efficient data collection strategies. In particular, we demonstrate that if annotations are collected in video frames, their efficacy can be multiplied for free by using motion cues. To explore this idea, we introduce DensePose-Track, a dataset of videos where selected frames are annotated in the traditional DensePose manner. Then, building on geometric properties of the DensePose mapping, we use the video dynamic to propagate ground-truth annotations in time as well as to learn from Siamese equivariance constraints. Having performed exhaustive empirical evaluation of various data annotation and learning strategies, we demonstrate that doing so can deliver significantly improved pose estimation results over strong baselines. However, despite what is suggested by some recent works, we show that merely synthesizing motion patterns by applying geometric transformations to isolated frames is significantly less effective, and that motion cues help much more when they are extracted from videos.
{"title":"Slim DensePose: Thrifty Learning From Sparse Annotations and Motion Cues","authors":"N. Neverova, James Thewlis, R. Güler, Iasonas Kokkinos, A. Vedaldi","doi":"10.1109/CVPR.2019.01117","DOIUrl":"https://doi.org/10.1109/CVPR.2019.01117","url":null,"abstract":"DensePose supersedes traditional landmark detectors by densely mapping image pixels to body surface coordinates. This power, however, comes at a greatly increased annotation cost, as supervising the model requires to manually label hundreds of points per pose instance. In this work, we thus seek methods to significantly slim down the DensePose annotations, proposing more efficient data collection strategies. In particular, we demonstrate that if annotations are collected in video frames, their efficacy can be multiplied for free by using motion cues. To explore this idea, we introduce DensePose-Track, a dataset of videos where selected frames are annotated in the traditional DensePose manner. Then, building on geometric properties of the DensePose mapping, we use the video dynamic to propagate ground-truth annotations in time as well as to learn from Siamese equivariance constraints. Having performed exhaustive empirical evaluation of various data annotation and learning strategies, we demonstrate that doing so can deliver significantly improved pose estimation results over strong baselines. However, despite what is suggested by some recent works, we show that merely synthesizing motion patterns by applying geometric transformations to isolated frames is significantly less effective, and that motion cues help much more when they are extracted from videos.","PeriodicalId":6711,"journal":{"name":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"28 1","pages":"10907-10915"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75493737","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}
Face anti-spoofing is an important task in full-stack face applications including face detection, verification, and recognition. Previous approaches build models on datasets which do not simulate the real-world data well (e.g., small scale, insignificant variance, etc.). Existing models may rely on auxiliary information, which prevents these anti-spoofing solutions from generalizing well in practice. In this paper, we present a data collection solution along with a data synthesis technique to simulate digital medium-based face spoofing attacks, which can easily help us obtain a large amount of training data well reflecting the real-world scenarios. Through exploiting a novel Spatio-Temporal Anti-Spoof Network (STASN), we are able to push the performance on public face anti-spoofing datasets over state-of-the-art methods by a large margin. Since the proposed model can automatically attend to discriminative regions, it makes analyzing the behaviors of the network possible.We conduct extensive experiments and show that the proposed model can distinguish spoof faces by extracting features from a variety of regions to seek out subtle evidences such as borders, moire patterns, reflection artifacts, etc.
{"title":"Face Anti-Spoofing: Model Matters, so Does Data","authors":"Xiao Yang, Wenhan Luo, Linchao Bao, Yuan Gao, Dihong Gong, Shibao Zheng, Zhifeng Li, Wei Liu","doi":"10.1109/CVPR.2019.00362","DOIUrl":"https://doi.org/10.1109/CVPR.2019.00362","url":null,"abstract":"Face anti-spoofing is an important task in full-stack face applications including face detection, verification, and recognition. Previous approaches build models on datasets which do not simulate the real-world data well (e.g., small scale, insignificant variance, etc.). Existing models may rely on auxiliary information, which prevents these anti-spoofing solutions from generalizing well in practice. In this paper, we present a data collection solution along with a data synthesis technique to simulate digital medium-based face spoofing attacks, which can easily help us obtain a large amount of training data well reflecting the real-world scenarios. Through exploiting a novel Spatio-Temporal Anti-Spoof Network (STASN), we are able to push the performance on public face anti-spoofing datasets over state-of-the-art methods by a large margin. Since the proposed model can automatically attend to discriminative regions, it makes analyzing the behaviors of the network possible.We conduct extensive experiments and show that the proposed model can distinguish spoof faces by extracting features from a variety of regions to seek out subtle evidences such as borders, moire patterns, reflection artifacts, etc.","PeriodicalId":6711,"journal":{"name":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"140 1","pages":"3502-3511"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73971009","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 develop a framework for novelty detection (ND) methods relying on deep embeddings, either discriminative or generative, and also propose a novel framework for assessing their performance. While much progress was made recently in these approaches, it has been accompanied by certain limitations: most methods were tested on relatively simple problems (low resolution images / small number of classes) or involved non-public data; comparative performance has often proven inconclusive because of lacking statistical significance; and evaluation has generally been done on non-canonical problem sets of differing complexity, making apples-to-apples comparative performance evaluation difficult. This has led to a relative confusing state of affairs. We address these challenges via the following contributions: We make a proposal for a novel framework to measure the performance of novelty detection methods using a trade-space demonstrating performance (measured by ROCAUC) as a function of problem complexity. We also make several proposals to formally characterize problem complexity. We conduct experiments with problems of higher complexity (higher image resolution / number of classes). To this end we design several canonical datasets built from CIFAR-10 and ImageNet (IN-125) which we make available to perform future benchmarks for novelty detection as well as other related tasks including semantic zero/adaptive shot and unsupervised learning. Finally, we demonstrate, as one of the methods in our ND framework, a generative novelty detection method whose performance exceeds that of all recent best-in-class generative ND methods.
{"title":"Where's Wally Now? Deep Generative and Discriminative Embeddings for Novelty Detection","authors":"P. Burlina, Neil J. Joshi, I-J. Wang","doi":"10.1109/CVPR.2019.01177","DOIUrl":"https://doi.org/10.1109/CVPR.2019.01177","url":null,"abstract":"We develop a framework for novelty detection (ND) methods relying on deep embeddings, either discriminative or generative, and also propose a novel framework for assessing their performance. While much progress was made recently in these approaches, it has been accompanied by certain limitations: most methods were tested on relatively simple problems (low resolution images / small number of classes) or involved non-public data; comparative performance has often proven inconclusive because of lacking statistical significance; and evaluation has generally been done on non-canonical problem sets of differing complexity, making apples-to-apples comparative performance evaluation difficult. This has led to a relative confusing state of affairs. We address these challenges via the following contributions: We make a proposal for a novel framework to measure the performance of novelty detection methods using a trade-space demonstrating performance (measured by ROCAUC) as a function of problem complexity. We also make several proposals to formally characterize problem complexity. We conduct experiments with problems of higher complexity (higher image resolution / number of classes). To this end we design several canonical datasets built from CIFAR-10 and ImageNet (IN-125) which we make available to perform future benchmarks for novelty detection as well as other related tasks including semantic zero/adaptive shot and unsupervised learning. Finally, we demonstrate, as one of the methods in our ND framework, a generative novelty detection method whose performance exceeds that of all recent best-in-class generative ND methods.","PeriodicalId":6711,"journal":{"name":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"22 1","pages":"11499-11508"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74902313","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}
Junxuan Li, A. Robles-Kelly, Shaodi You, Y. Matsushita
Photometric stereo estimates the surface normal given a set of images acquired under different illumination conditions. To deal with diverse factors involved in the image formation process, recent photometric stereo methods demand a large number of images as input. We propose a method that can dramatically decrease the demands on the number of images by learning the most informative ones under different illumination conditions. To this end, we use a deep learning framework to automatically learn the critical illumination conditions required at input. Furthermore, we present an occlusion layer that can synthesize cast shadows, which effectively improves the estimation accuracy. We assess our method on challenging real-world conditions, where we outperform techniques elsewhere in the literature with a significantly reduced number of light conditions.
{"title":"Learning to Minify Photometric Stereo","authors":"Junxuan Li, A. Robles-Kelly, Shaodi You, Y. Matsushita","doi":"10.1109/CVPR.2019.00775","DOIUrl":"https://doi.org/10.1109/CVPR.2019.00775","url":null,"abstract":"Photometric stereo estimates the surface normal given a set of images acquired under different illumination conditions. To deal with diverse factors involved in the image formation process, recent photometric stereo methods demand a large number of images as input. We propose a method that can dramatically decrease the demands on the number of images by learning the most informative ones under different illumination conditions. To this end, we use a deep learning framework to automatically learn the critical illumination conditions required at input. Furthermore, we present an occlusion layer that can synthesize cast shadows, which effectively improves the estimation accuracy. We assess our method on challenging real-world conditions, where we outperform techniques elsewhere in the literature with a significantly reduced number of light conditions.","PeriodicalId":6711,"journal":{"name":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"416 1","pages":"7560-7568"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76475852","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}
Hundreds of materials, such as drugs, explosives, makeup, food additives, are in the form of powder. Recognizing such powders is important for security checks, criminal identification, drug control, and quality assessment. However, powder recognition has drawn little attention in the computer vision community. Powders are hard to distinguish: they are amorphous, appear matte, have little color or texture variation and blend with surfaces they are deposited on in complex ways. To address these challenges, we present the first comprehensive dataset and approach for powder recognition using multi-spectral imaging. By using Shortwave Infrared (SWIR) multi-spectral imaging together with visible light (RGB) and Near Infrared (NIR), powders can be discriminated with reasonable accuracy. We present a method to select discriminative spectral bands to significantly reduce acquisition time while improving recognition accuracy. We propose a blending model to synthesize images of powders of various thickness deposited on a wide range of surfaces. Incorporating band selection and image synthesis, we conduct fine-grained recognition of 100 powders on complex backgrounds, and achieve 60%~70% accuracy on recognition with known powder location, and over 40% mean IoU without known location.
{"title":"Multispectral Imaging for Fine-Grained Recognition of Powders on Complex Backgrounds","authors":"Tiancheng Zhi, B. Pires, M. Hebert, S. Narasimhan","doi":"10.1109/CVPR.2019.00890","DOIUrl":"https://doi.org/10.1109/CVPR.2019.00890","url":null,"abstract":"Hundreds of materials, such as drugs, explosives, makeup, food additives, are in the form of powder. Recognizing such powders is important for security checks, criminal identification, drug control, and quality assessment. However, powder recognition has drawn little attention in the computer vision community. Powders are hard to distinguish: they are amorphous, appear matte, have little color or texture variation and blend with surfaces they are deposited on in complex ways. To address these challenges, we present the first comprehensive dataset and approach for powder recognition using multi-spectral imaging. By using Shortwave Infrared (SWIR) multi-spectral imaging together with visible light (RGB) and Near Infrared (NIR), powders can be discriminated with reasonable accuracy. We present a method to select discriminative spectral bands to significantly reduce acquisition time while improving recognition accuracy. We propose a blending model to synthesize images of powders of various thickness deposited on a wide range of surfaces. Incorporating band selection and image synthesis, we conduct fine-grained recognition of 100 powders on complex backgrounds, and achieve 60%~70% accuracy on recognition with known powder location, and over 40% mean IoU without known location.","PeriodicalId":6711,"journal":{"name":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"82 1","pages":"8691-8700"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80143971","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}
Jifei Song, Yongxin Yang, Yi-Zhe Song, T. Xiang, Timothy M. Hospedales
We aim to learn a domain generalizable person re-identification (ReID) model. When such a model is trained on a set of source domains (ReID datasets collected from different camera networks), it can be directly applied to any new unseen dataset for effective ReID without any model updating. Despite its practical value in real-world deployments, generalizable ReID has seldom been studied. In this work, a novel deep ReID model termed Domain-Invariant Mapping Network (DIMN) is proposed. DIMN is designed to learn a mapping between a person image and its identity classifier, i.e., it produces a classifier using a single shot. To make the model domain-invariant, we follow a meta-learning pipeline and sample a subset of source domain training tasks during each training episode. However, the model is significantly different from conventional meta-learning methods in that: (1) no model updating is required for the target domain, (2) different training tasks share a memory bank for maintaining both scalability and discrimination ability, and (3) it can be used to match an arbitrary number of identities in a target domain. Extensive experiments on a newly proposed large-scale ReID domain generalization benchmark show that our DIMN significantly outperforms alternative domain generalization or meta-learning methods.
{"title":"Generalizable Person Re-Identification by Domain-Invariant Mapping Network","authors":"Jifei Song, Yongxin Yang, Yi-Zhe Song, T. Xiang, Timothy M. Hospedales","doi":"10.1109/CVPR.2019.00081","DOIUrl":"https://doi.org/10.1109/CVPR.2019.00081","url":null,"abstract":"We aim to learn a domain generalizable person re-identification (ReID) model. When such a model is trained on a set of source domains (ReID datasets collected from different camera networks), it can be directly applied to any new unseen dataset for effective ReID without any model updating. Despite its practical value in real-world deployments, generalizable ReID has seldom been studied. In this work, a novel deep ReID model termed Domain-Invariant Mapping Network (DIMN) is proposed. DIMN is designed to learn a mapping between a person image and its identity classifier, i.e., it produces a classifier using a single shot. To make the model domain-invariant, we follow a meta-learning pipeline and sample a subset of source domain training tasks during each training episode. However, the model is significantly different from conventional meta-learning methods in that: (1) no model updating is required for the target domain, (2) different training tasks share a memory bank for maintaining both scalability and discrimination ability, and (3) it can be used to match an arbitrary number of identities in a target domain. Extensive experiments on a newly proposed large-scale ReID domain generalization benchmark show that our DIMN significantly outperforms alternative domain generalization or meta-learning methods.","PeriodicalId":6711,"journal":{"name":"2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"18 1","pages":"719-728"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81829094","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：