Liangqiong Qu, Jiandong Tian, Shengfeng He, Yandong Tang, Rynson W. H. Lau
Shadow removal is a challenging task as it requires the detection/annotation of shadows as well as semantic understanding of the scene. In this paper, we propose an automatic and end-to-end deep neural network (DeshadowNet) to tackle these problems in a unified manner. DeshadowNet is designed with a multi-context architecture, where the output shadow matte is predicted by embedding information from three different perspectives. The first global network extracts shadow features from a global view. Two levels of features are derived from the global network and transferred to two parallel networks. While one extracts the appearance of the input image, the other one involves semantic understanding for final prediction. These two complementary networks generate multi-context features to obtain the shadow matte with fine local details. To evaluate the performance of the proposed method, we construct the first large scale benchmark with 3088 image pairs. Extensive experiments on two publicly available benchmarks and our large-scale benchmark show that the proposed method performs favorably against several state-of-the-art methods.
{"title":"DeshadowNet: A Multi-context Embedding Deep Network for Shadow Removal","authors":"Liangqiong Qu, Jiandong Tian, Shengfeng He, Yandong Tang, Rynson W. H. Lau","doi":"10.1109/CVPR.2017.248","DOIUrl":"https://doi.org/10.1109/CVPR.2017.248","url":null,"abstract":"Shadow removal is a challenging task as it requires the detection/annotation of shadows as well as semantic understanding of the scene. In this paper, we propose an automatic and end-to-end deep neural network (DeshadowNet) to tackle these problems in a unified manner. DeshadowNet is designed with a multi-context architecture, where the output shadow matte is predicted by embedding information from three different perspectives. The first global network extracts shadow features from a global view. Two levels of features are derived from the global network and transferred to two parallel networks. While one extracts the appearance of the input image, the other one involves semantic understanding for final prediction. These two complementary networks generate multi-context features to obtain the shadow matte with fine local details. To evaluate the performance of the proposed method, we construct the first large scale benchmark with 3088 image pairs. Extensive experiments on two publicly available benchmarks and our large-scale benchmark show that the proposed method performs favorably against several state-of-the-art methods.","PeriodicalId":6631,"journal":{"name":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"60 1","pages":"2308-2316"},"PeriodicalIF":0.0,"publicationDate":"2017-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81049382","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}
Recent low-rank based matrix/tensor recovery methods have been widely explored in multispectral images (MSI) denoising. These methods, however, ignore the difference of the intrinsic structure correlation along spatial sparsity, spectral correlation and non-local self-similarity mode. In this paper, we go further by giving a detailed analysis about the rank properties both in matrix and tensor cases, and figure out the non-local self-similarity is the key ingredient, while the low-rank assumption of others may not hold. This motivates us to design a simple yet effective unidirectional low-rank tensor recovery model that is capable of truthfully capturing the intrinsic structure correlation with reduced computational burden. However, the low-rank models suffer from the ringing artifacts, due to the aggregation of overlapped patches/cubics. While previous methods resort to spatial information, we offer a new perspective by utilizing the exclusively spectral information in MSIs to address the issue. The analysis-based hyper-Laplacian prior is introduced to model the global spectral structures, so as to indirectly alleviate the ringing artifacts in spatial domain. The advantages of the proposed method over the existing ones are multi-fold: more reasonably structure correlation representability, less processing time, and less artifacts in the overlapped regions. The proposed method is extensively evaluated on several benchmarks, and significantly outperforms state-of-the-art MSI denoising methods.
{"title":"Hyper-Laplacian Regularized Unidirectional Low-Rank Tensor Recovery for Multispectral Image Denoising","authors":"Yi Chang, Luxin Yan, Sheng Zhong","doi":"10.1109/CVPR.2017.625","DOIUrl":"https://doi.org/10.1109/CVPR.2017.625","url":null,"abstract":"Recent low-rank based matrix/tensor recovery methods have been widely explored in multispectral images (MSI) denoising. These methods, however, ignore the difference of the intrinsic structure correlation along spatial sparsity, spectral correlation and non-local self-similarity mode. In this paper, we go further by giving a detailed analysis about the rank properties both in matrix and tensor cases, and figure out the non-local self-similarity is the key ingredient, while the low-rank assumption of others may not hold. This motivates us to design a simple yet effective unidirectional low-rank tensor recovery model that is capable of truthfully capturing the intrinsic structure correlation with reduced computational burden. However, the low-rank models suffer from the ringing artifacts, due to the aggregation of overlapped patches/cubics. While previous methods resort to spatial information, we offer a new perspective by utilizing the exclusively spectral information in MSIs to address the issue. The analysis-based hyper-Laplacian prior is introduced to model the global spectral structures, so as to indirectly alleviate the ringing artifacts in spatial domain. The advantages of the proposed method over the existing ones are multi-fold: more reasonably structure correlation representability, less processing time, and less artifacts in the overlapped regions. The proposed method is extensively evaluated on several benchmarks, and significantly outperforms state-of-the-art MSI denoising methods.","PeriodicalId":6631,"journal":{"name":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"32 1","pages":"5901-5909"},"PeriodicalIF":0.0,"publicationDate":"2017-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84796285","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}
Recently, plugging trainable structural layers into deep convolutional neural networks (CNNs) as image representations has made promising progress. However, there has been little work on inserting parametric probability distributions, which can effectively model feature statistics, into deep CNNs in an end-to-end manner. This paper proposes a Global Gaussian Distribution embedding Network (G2DeNet) to take a step towards addressing this problem. The core of G2DeNet is a novel trainable layer of a global Gaussian as an image representation plugged into deep CNNs for end-to-end learning. The challenge is that the proposed layer involves Gaussian distributions whose space is not a linear space, which makes its forward and backward propagations be non-intuitive and non-trivial. To tackle this issue, we employ a Gaussian embedding strategy which respects the structures of both Riemannian manifold and smooth group of Gaussians. Based on this strategy, we construct the proposed global Gaussian embedding layer and decompose it into two sub-layers: the matrix partition sub-layer decoupling the mean vector and covariance matrix entangled in the embedding matrix, and the square-rooted, symmetric positive definite matrix sub-layer. In this way, we can derive the partial derivatives associated with the proposed structural layer and thus allow backpropagation of gradients. Experimental results on large scale region classification and fine-grained recognition tasks show that G2DeNet is superior to its counterparts, capable of achieving state-of-the-art performance.
{"title":"G2DeNet: Global Gaussian Distribution Embedding Network and Its Application to Visual Recognition","authors":"Qilong Wang, P. Li, Lei Zhang","doi":"10.1109/CVPR.2017.689","DOIUrl":"https://doi.org/10.1109/CVPR.2017.689","url":null,"abstract":"Recently, plugging trainable structural layers into deep convolutional neural networks (CNNs) as image representations has made promising progress. However, there has been little work on inserting parametric probability distributions, which can effectively model feature statistics, into deep CNNs in an end-to-end manner. This paper proposes a Global Gaussian Distribution embedding Network (G2DeNet) to take a step towards addressing this problem. The core of G2DeNet is a novel trainable layer of a global Gaussian as an image representation plugged into deep CNNs for end-to-end learning. The challenge is that the proposed layer involves Gaussian distributions whose space is not a linear space, which makes its forward and backward propagations be non-intuitive and non-trivial. To tackle this issue, we employ a Gaussian embedding strategy which respects the structures of both Riemannian manifold and smooth group of Gaussians. Based on this strategy, we construct the proposed global Gaussian embedding layer and decompose it into two sub-layers: the matrix partition sub-layer decoupling the mean vector and covariance matrix entangled in the embedding matrix, and the square-rooted, symmetric positive definite matrix sub-layer. In this way, we can derive the partial derivatives associated with the proposed structural layer and thus allow backpropagation of gradients. Experimental results on large scale region classification and fine-grained recognition tasks show that G2DeNet is superior to its counterparts, capable of achieving state-of-the-art performance.","PeriodicalId":6631,"journal":{"name":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"198 1","pages":"6507-6516"},"PeriodicalIF":0.0,"publicationDate":"2017-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90359214","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 fine-grained categories (e.g., bird species) is difficult due to the challenges of discriminative region localization and fine-grained feature learning. Existing approaches predominantly solve these challenges independently, while neglecting the fact that region detection and fine-grained feature learning are mutually correlated and thus can reinforce each other. In this paper, we propose a novel recurrent attention convolutional neural network (RA-CNN) which recursively learns discriminative region attention and region-based feature representation at multiple scales in a mutual reinforced way. The learning at each scale consists of a classification sub-network and an attention proposal sub-network (APN). The APN starts from full images, and iteratively generates region attention from coarse to fine by taking previous prediction as a reference, while the finer scale network takes as input an amplified attended region from previous scale in a recurrent way. The proposed RA-CNN is optimized by an intra-scale classification loss and an inter-scale ranking loss, to mutually learn accurate region attention and fine-grained representation. RA-CNN does not need bounding box/part annotations and can be trained end-to-end. We conduct comprehensive experiments and show that RA-CNN achieves the best performance in three fine-grained tasks, with relative accuracy gains of 3.3%, 3.7%, 3.8%, on CUB Birds, Stanford Dogs and Stanford Cars, respectively.
{"title":"Look Closer to See Better: Recurrent Attention Convolutional Neural Network for Fine-Grained Image Recognition","authors":"Jianlong Fu, Heliang Zheng, Tao Mei","doi":"10.1109/CVPR.2017.476","DOIUrl":"https://doi.org/10.1109/CVPR.2017.476","url":null,"abstract":"Recognizing fine-grained categories (e.g., bird species) is difficult due to the challenges of discriminative region localization and fine-grained feature learning. Existing approaches predominantly solve these challenges independently, while neglecting the fact that region detection and fine-grained feature learning are mutually correlated and thus can reinforce each other. In this paper, we propose a novel recurrent attention convolutional neural network (RA-CNN) which recursively learns discriminative region attention and region-based feature representation at multiple scales in a mutual reinforced way. The learning at each scale consists of a classification sub-network and an attention proposal sub-network (APN). The APN starts from full images, and iteratively generates region attention from coarse to fine by taking previous prediction as a reference, while the finer scale network takes as input an amplified attended region from previous scale in a recurrent way. The proposed RA-CNN is optimized by an intra-scale classification loss and an inter-scale ranking loss, to mutually learn accurate region attention and fine-grained representation. RA-CNN does not need bounding box/part annotations and can be trained end-to-end. We conduct comprehensive experiments and show that RA-CNN achieves the best performance in three fine-grained tasks, with relative accuracy gains of 3.3%, 3.7%, 3.8%, on CUB Birds, Stanford Dogs and Stanford Cars, respectively.","PeriodicalId":6631,"journal":{"name":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"15 1","pages":"4476-4484"},"PeriodicalIF":0.0,"publicationDate":"2017-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73415819","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}
Yanyang Yan, Wenqi Ren, Yuanfang Guo, Rui Wang, Xiaochun Cao
Camera motion introduces motion blur, affecting many computer vision tasks. Dark Channel Prior (DCP) helps the blind deblurring on scenes including natural, face, text, and low-illumination images. However, it has limitations and is less likely to support the kernel estimation while bright pixels dominate the input image. We observe that the bright pixels in the clear images are not likely to be bright after the blur process. Based on this observation, we first illustrate this phenomenon mathematically and define it as the Bright Channel Prior (BCP). Then, we propose a technique for deblurring such images which elevates the performance of existing motion deblurring algorithms. The proposed method takes advantage of both Bright and Dark Channel Prior. This joint prior is named as extreme channels prior and is crucial for achieving efficient restorations by leveraging both the bright and dark information. Extensive experimental results demonstrate that the proposed method is more robust and performs favorably against the state-of-the-art image deblurring methods on both synthesized and natural images.
{"title":"Image Deblurring via Extreme Channels Prior","authors":"Yanyang Yan, Wenqi Ren, Yuanfang Guo, Rui Wang, Xiaochun Cao","doi":"10.1109/CVPR.2017.738","DOIUrl":"https://doi.org/10.1109/CVPR.2017.738","url":null,"abstract":"Camera motion introduces motion blur, affecting many computer vision tasks. Dark Channel Prior (DCP) helps the blind deblurring on scenes including natural, face, text, and low-illumination images. However, it has limitations and is less likely to support the kernel estimation while bright pixels dominate the input image. We observe that the bright pixels in the clear images are not likely to be bright after the blur process. Based on this observation, we first illustrate this phenomenon mathematically and define it as the Bright Channel Prior (BCP). Then, we propose a technique for deblurring such images which elevates the performance of existing motion deblurring algorithms. The proposed method takes advantage of both Bright and Dark Channel Prior. This joint prior is named as extreme channels prior and is crucial for achieving efficient restorations by leveraging both the bright and dark information. Extensive experimental results demonstrate that the proposed method is more robust and performs favorably against the state-of-the-art image deblurring methods on both synthesized and natural images.","PeriodicalId":6631,"journal":{"name":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"5 1","pages":"6978-6986"},"PeriodicalIF":0.0,"publicationDate":"2017-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79269453","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}
S. Buch, Victor Escorcia, Chuanqi Shen, Bernard Ghanem, Juan Carlos Niebles
Our paper presents a new approach for temporal detection of human actions in long, untrimmed video sequences. We introduce Single-Stream Temporal Action Proposals (SST), a new effective and efficient deep architecture for the generation of temporal action proposals. Our network can run continuously in a single stream over very long input video sequences, without the need to divide input into short overlapping clips or temporal windows for batch processing. We demonstrate empirically that our model outperforms the state-of-the-art on the task of temporal action proposal generation, while achieving some of the fastest processing speeds in the literature. Finally, we demonstrate that using SST proposals in conjunction with existing action classifiers results in improved state-of-the-art temporal action detection performance.
{"title":"SST: Single-Stream Temporal Action Proposals","authors":"S. Buch, Victor Escorcia, Chuanqi Shen, Bernard Ghanem, Juan Carlos Niebles","doi":"10.1109/CVPR.2017.675","DOIUrl":"https://doi.org/10.1109/CVPR.2017.675","url":null,"abstract":"Our paper presents a new approach for temporal detection of human actions in long, untrimmed video sequences. We introduce Single-Stream Temporal Action Proposals (SST), a new effective and efficient deep architecture for the generation of temporal action proposals. Our network can run continuously in a single stream over very long input video sequences, without the need to divide input into short overlapping clips or temporal windows for batch processing. We demonstrate empirically that our model outperforms the state-of-the-art on the task of temporal action proposal generation, while achieving some of the fastest processing speeds in the literature. Finally, we demonstrate that using SST proposals in conjunction with existing action classifiers results in improved state-of-the-art temporal action detection performance.","PeriodicalId":6631,"journal":{"name":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"27 1","pages":"6373-6382"},"PeriodicalIF":0.0,"publicationDate":"2017-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81896170","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}
George Trigeorgis, Patrick Snape, Iasonas Kokkinos, S. Zafeiriou
In this work we pursue a data-driven approach to the problem of estimating surface normals from a single intensity image, focusing in particular on human faces. We introduce new methods to exploit the currently available facial databases for dataset construction and tailor a deep convolutional neural network to the task of estimating facial surface normals in-the-wild. We train a fully convolutional network that can accurately recover facial normals from images including a challenging variety of expressions and facial poses. We compare against state-of-the-art face Shape-from-Shading and 3D reconstruction techniques and show that the proposed network can recover substantially more accurate and realistic normals. Furthermore, in contrast to other existing face-specific surface recovery methods, we do not require the solving of an explicit alignment step due to the fully convolutional nature of our network.
{"title":"Face Normals \"In-the-Wild\" Using Fully Convolutional Networks","authors":"George Trigeorgis, Patrick Snape, Iasonas Kokkinos, S. Zafeiriou","doi":"10.1109/CVPR.2017.44","DOIUrl":"https://doi.org/10.1109/CVPR.2017.44","url":null,"abstract":"In this work we pursue a data-driven approach to the problem of estimating surface normals from a single intensity image, focusing in particular on human faces. We introduce new methods to exploit the currently available facial databases for dataset construction and tailor a deep convolutional neural network to the task of estimating facial surface normals in-the-wild. We train a fully convolutional network that can accurately recover facial normals from images including a challenging variety of expressions and facial poses. We compare against state-of-the-art face Shape-from-Shading and 3D reconstruction techniques and show that the proposed network can recover substantially more accurate and realistic normals. Furthermore, in contrast to other existing face-specific surface recovery methods, we do not require the solving of an explicit alignment step due to the fully convolutional nature of our network.","PeriodicalId":6631,"journal":{"name":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"7 1","pages":"340-349"},"PeriodicalIF":0.0,"publicationDate":"2017-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84185172","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}
Recent advances in the joint processing of images have certainly shown its advantages over the individual processing. Different from the existing works geared towards co-segmentation or co-localization, in this paper, we explore a new joint processing topic: co-skeletonization, which is defined as joint skeleton extraction of common objects in a set of semantically similar images. Object skeletonization in real world images is a challenging problem, because there is no prior knowledge of the objects shape if we consider only a single image. This motivates us to resort to the idea of object co-skeletonization hoping that the commonness prior existing across the similar images may help, just as it does for other joint processing problems such as co-segmentation. Noting that skeleton can provide good scribbles for segmentation, and skeletonization, in turn, needs good segmentation, we propose a coupled framework for co-skeletonization and co-segmentation tasks so that they are well informed by each other, and benefit each other synergistically. Since it is a new problem, we also construct a benchmark dataset for the co-skeletonization task. Extensive experiments demonstrate that proposed method achieves very competitive results.
{"title":"Object Co-skeletonization with Co-segmentation","authors":"Koteswar Rao Jerripothula, Jianfei Cai, Jiangbo Lu, Junsong Yuan","doi":"10.1109/CVPR.2017.413","DOIUrl":"https://doi.org/10.1109/CVPR.2017.413","url":null,"abstract":"Recent advances in the joint processing of images have certainly shown its advantages over the individual processing. Different from the existing works geared towards co-segmentation or co-localization, in this paper, we explore a new joint processing topic: co-skeletonization, which is defined as joint skeleton extraction of common objects in a set of semantically similar images. Object skeletonization in real world images is a challenging problem, because there is no prior knowledge of the objects shape if we consider only a single image. This motivates us to resort to the idea of object co-skeletonization hoping that the commonness prior existing across the similar images may help, just as it does for other joint processing problems such as co-segmentation. Noting that skeleton can provide good scribbles for segmentation, and skeletonization, in turn, needs good segmentation, we propose a coupled framework for co-skeletonization and co-segmentation tasks so that they are well informed by each other, and benefit each other synergistically. Since it is a new problem, we also construct a benchmark dataset for the co-skeletonization task. Extensive experiments demonstrate that proposed method achieves very competitive results.","PeriodicalId":6631,"journal":{"name":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"41 1","pages":"3881-3889"},"PeriodicalIF":0.0,"publicationDate":"2017-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89260881","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}
Chuang Gan, Zhe Gan, Xiaodong He, Jianfeng Gao, L. Deng
We propose a novel framework named StyleNet to address the task of generating attractive captions for images and videos with different styles. To this end, we devise a novel model component, named factored LSTM, which automatically distills the style factors in the monolingual text corpus. Then at runtime, we can explicitly control the style in the caption generation process so as to produce attractive visual captions with the desired style. Our approach achieves this goal by leveraging two sets of data: 1) factual image/video-caption paired data, and 2) stylized monolingual text data (e.g., romantic and humorous sentences). We show experimentally that StyleNet outperforms existing approaches for generating visual captions with different styles, measured in both automatic and human evaluation metrics on the newly collected FlickrStyle10K image caption dataset, which contains 10K Flickr images with corresponding humorous and romantic captions.
{"title":"StyleNet: Generating Attractive Visual Captions with Styles","authors":"Chuang Gan, Zhe Gan, Xiaodong He, Jianfeng Gao, L. Deng","doi":"10.1109/CVPR.2017.108","DOIUrl":"https://doi.org/10.1109/CVPR.2017.108","url":null,"abstract":"We propose a novel framework named StyleNet to address the task of generating attractive captions for images and videos with different styles. To this end, we devise a novel model component, named factored LSTM, which automatically distills the style factors in the monolingual text corpus. Then at runtime, we can explicitly control the style in the caption generation process so as to produce attractive visual captions with the desired style. Our approach achieves this goal by leveraging two sets of data: 1) factual image/video-caption paired data, and 2) stylized monolingual text data (e.g., romantic and humorous sentences). We show experimentally that StyleNet outperforms existing approaches for generating visual captions with different styles, measured in both automatic and human evaluation metrics on the newly collected FlickrStyle10K image caption dataset, which contains 10K Flickr images with corresponding humorous and romantic captions.","PeriodicalId":6631,"journal":{"name":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"11 1","pages":"955-964"},"PeriodicalIF":0.0,"publicationDate":"2017-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86353250","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}
This paper deals with deep neural networks for predicting accurate dense disparity map with Semi-global matching (SGM). SGM is a widely used regularization method for real scenes because of its high accuracy and fast computation speed. Even though SGM can obtain accurate results, tuning of SGMs penalty-parameters, which control a smoothness and discontinuity of a disparity map, is uneasy and empirical methods have been proposed. We propose a learning based penalties estimation method, which we call SGM-Nets that consist of Convolutional Neural Networks. A small image patch and its position are input into SGMNets to predict the penalties for the 3D object structures. In order to train the networks, we introduce a novel loss function which is able to use sparsely annotated disparity maps such as captured by a LiDAR sensor in real environments. Moreover, we propose a novel SGM parameterization, which deploys different penalties depending on either positive or negative disparity changes in order to represent the object structures more discriminatively. Our SGM-Nets outperformed state of the art accuracy on KITTI benchmark datasets.
{"title":"SGM-Nets: Semi-Global Matching with Neural Networks","authors":"A. Seki, M. Pollefeys","doi":"10.1109/CVPR.2017.703","DOIUrl":"https://doi.org/10.1109/CVPR.2017.703","url":null,"abstract":"This paper deals with deep neural networks for predicting accurate dense disparity map with Semi-global matching (SGM). SGM is a widely used regularization method for real scenes because of its high accuracy and fast computation speed. Even though SGM can obtain accurate results, tuning of SGMs penalty-parameters, which control a smoothness and discontinuity of a disparity map, is uneasy and empirical methods have been proposed. We propose a learning based penalties estimation method, which we call SGM-Nets that consist of Convolutional Neural Networks. A small image patch and its position are input into SGMNets to predict the penalties for the 3D object structures. In order to train the networks, we introduce a novel loss function which is able to use sparsely annotated disparity maps such as captured by a LiDAR sensor in real environments. Moreover, we propose a novel SGM parameterization, which deploys different penalties depending on either positive or negative disparity changes in order to represent the object structures more discriminatively. Our SGM-Nets outperformed state of the art accuracy on KITTI benchmark datasets.","PeriodicalId":6631,"journal":{"name":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"39 1","pages":"6640-6649"},"PeriodicalIF":0.0,"publicationDate":"2017-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79004005","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：