Foreground object segmentation is a critical step for many image analysis tasks. While automated methods can produce high-quality results, their failures disappoint users in need of practical solutions. We propose a resource allocation framework for predicting how best to allocate a fixed budget of human annotation effort in order to collect higher quality segmentations for a given batch of images and automated methods. The framework is based on a proposed prediction module that estimates the quality of given algorithm-drawn segmentations. We demonstrate the value of the framework for two novel tasks related to "pulling the plug" on computer and human annotators. Specifically, we implement two systems that automatically decide, for a batch of images, when to replace 1) humans with computers to create coarse segmentations required to initialize segmentation tools and 2) computers with humans to create final, fine-grained segmentations. Experiments demonstrate the advantage of relying on a mix of human and computer efforts over relying on either resource alone for segmenting objects in three diverse datasets representing visible, phase contrast microscopy, and fluorescence microscopy images.
{"title":"Pull the Plug? Predicting If Computers or Humans Should Segment Images","authors":"D. Gurari, S. Jain, Margrit Betke, K. Grauman","doi":"10.1109/CVPR.2016.48","DOIUrl":"https://doi.org/10.1109/CVPR.2016.48","url":null,"abstract":"Foreground object segmentation is a critical step for many image analysis tasks. While automated methods can produce high-quality results, their failures disappoint users in need of practical solutions. We propose a resource allocation framework for predicting how best to allocate a fixed budget of human annotation effort in order to collect higher quality segmentations for a given batch of images and automated methods. The framework is based on a proposed prediction module that estimates the quality of given algorithm-drawn segmentations. We demonstrate the value of the framework for two novel tasks related to \"pulling the plug\" on computer and human annotators. Specifically, we implement two systems that automatically decide, for a batch of images, when to replace 1) humans with computers to create coarse segmentations required to initialize segmentation tools and 2) computers with humans to create final, fine-grained segmentations. Experiments demonstrate the advantage of relying on a mix of human and computer efforts over relying on either resource alone for segmenting objects in three diverse datasets representing visible, phase contrast microscopy, and fluorescence microscopy images.","PeriodicalId":6515,"journal":{"name":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"16 1","pages":"382-391"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91022843","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}
Hongye Liu, Yonghong Tian, Yaowei Wang, Lu Pang, Tiejun Huang
The growing explosion in the use of surveillance cameras in public security highlights the importance of vehicle search from a large-scale image or video database. However, compared with person re-identification or face recognition, vehicle search problem has long been neglected by researchers in vision community. This paper focuses on an interesting but challenging problem, vehicle re-identification (a.k.a precise vehicle search). We propose a Deep Relative Distance Learning (DRDL) method which exploits a two-branch deep convolutional network to project raw vehicle images into an Euclidean space where distance can be directly used to measure the similarity of arbitrary two vehicles. To further facilitate the future research on this problem, we also present a carefully-organized largescale image database "VehicleID", which includes multiple images of the same vehicle captured by different realworld cameras in a city. We evaluate our DRDL method on our VehicleID dataset and another recently-released vehicle model classification dataset "CompCars" in three sets of experiments: vehicle re-identification, vehicle model verification and vehicle retrieval. Experimental results show that our method can achieve promising results and outperforms several state-of-the-art approaches.
{"title":"Deep Relative Distance Learning: Tell the Difference between Similar Vehicles","authors":"Hongye Liu, Yonghong Tian, Yaowei Wang, Lu Pang, Tiejun Huang","doi":"10.1109/CVPR.2016.238","DOIUrl":"https://doi.org/10.1109/CVPR.2016.238","url":null,"abstract":"The growing explosion in the use of surveillance cameras in public security highlights the importance of vehicle search from a large-scale image or video database. However, compared with person re-identification or face recognition, vehicle search problem has long been neglected by researchers in vision community. This paper focuses on an interesting but challenging problem, vehicle re-identification (a.k.a precise vehicle search). We propose a Deep Relative Distance Learning (DRDL) method which exploits a two-branch deep convolutional network to project raw vehicle images into an Euclidean space where distance can be directly used to measure the similarity of arbitrary two vehicles. To further facilitate the future research on this problem, we also present a carefully-organized largescale image database \"VehicleID\", which includes multiple images of the same vehicle captured by different realworld cameras in a city. We evaluate our DRDL method on our VehicleID dataset and another recently-released vehicle model classification dataset \"CompCars\" in three sets of experiments: vehicle re-identification, vehicle model verification and vehicle retrieval. Experimental results show that our method can achieve promising results and outperforms several state-of-the-art approaches.","PeriodicalId":6515,"journal":{"name":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"10 1","pages":"2167-2175"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88931445","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}
Image segmentation is a key component in many computer vision systems, and it is recovering a prominent spot in the literature as methods improve and overcome their limitations. The outputs of most recent algorithms are in the form of a hierarchical segmentation, which provides segmentation at different scales in a single tree-like structure. Commonly, these hierarchical methods start from some low-level features, and are not aware of the scale information of the different regions in them. As such, one might need to work on many different levels of the hierarchy to find the objects in the scene. This work tries to modify the existing hierarchical algorithm by improving their alignment, that is, by trying to modify the depth of the regions in the tree to better couple depth and scale. To do so, we first train a regressor to predict the scale of regions using mid-level features. We then define the anchor slice as the set of regions that better balance between over-segmentation and under-segmentation. The output of our method is an improved hierarchy, re-aligned by the anchor slice. To demonstrate the power of our method, we perform comprehensive experiments, which show that our method, as a post-processing step, can significantly improve the quality of the hierarchical segmentation representations, and ease the usage of hierarchical image segmentation to high-level vision tasks such as object segmentation. We also prove that the improvement generalizes well across different algorithms and datasets, with a low computational cost.
{"title":"Scale-Aware Alignment of Hierarchical Image Segmentation","authors":"Yuhua Chen, Dengxin Dai, J. Pont-Tuset, L. Gool","doi":"10.1109/CVPR.2016.46","DOIUrl":"https://doi.org/10.1109/CVPR.2016.46","url":null,"abstract":"Image segmentation is a key component in many computer vision systems, and it is recovering a prominent spot in the literature as methods improve and overcome their limitations. The outputs of most recent algorithms are in the form of a hierarchical segmentation, which provides segmentation at different scales in a single tree-like structure. Commonly, these hierarchical methods start from some low-level features, and are not aware of the scale information of the different regions in them. As such, one might need to work on many different levels of the hierarchy to find the objects in the scene. This work tries to modify the existing hierarchical algorithm by improving their alignment, that is, by trying to modify the depth of the regions in the tree to better couple depth and scale. To do so, we first train a regressor to predict the scale of regions using mid-level features. We then define the anchor slice as the set of regions that better balance between over-segmentation and under-segmentation. The output of our method is an improved hierarchy, re-aligned by the anchor slice. To demonstrate the power of our method, we perform comprehensive experiments, which show that our method, as a post-processing step, can significantly improve the quality of the hierarchical segmentation representations, and ease the usage of hierarchical image segmentation to high-level vision tasks such as object segmentation. We also prove that the improvement generalizes well across different algorithms and datasets, with a low computational cost.","PeriodicalId":6515,"journal":{"name":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"45 1","pages":"364-372"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90800903","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 improve training of temporal deep models to better learn activity progression for activity detection and early detection tasks. Conventionally, when training a Recurrent Neural Network, specifically a Long Short Term Memory (LSTM) model, the training loss only considers classification error. However, we argue that the detection score of the correct activity category, or the detection score margin between the correct and incorrect categories, should be monotonically non-decreasing as the model observes more of the activity. We design novel ranking losses that directly penalize the model on violation of such monotonicities, which are used together with classification loss in training of LSTM models. Evaluation on ActivityNet shows significant benefits of the proposed ranking losses in both activity detection and early detection tasks.
{"title":"Learning Activity Progression in LSTMs for Activity Detection and Early Detection","authors":"Shugao Ma, L. Sigal, S. Sclaroff","doi":"10.1109/CVPR.2016.214","DOIUrl":"https://doi.org/10.1109/CVPR.2016.214","url":null,"abstract":"In this work we improve training of temporal deep models to better learn activity progression for activity detection and early detection tasks. Conventionally, when training a Recurrent Neural Network, specifically a Long Short Term Memory (LSTM) model, the training loss only considers classification error. However, we argue that the detection score of the correct activity category, or the detection score margin between the correct and incorrect categories, should be monotonically non-decreasing as the model observes more of the activity. We design novel ranking losses that directly penalize the model on violation of such monotonicities, which are used together with classification loss in training of LSTM models. Evaluation on ActivityNet shows significant benefits of the proposed ranking losses in both activity detection and early detection tasks.","PeriodicalId":6515,"journal":{"name":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"5 1","pages":"1942-1950"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89692520","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}
Many non-rigid 3D structures are not modelled well through a low-rank subspace assumption. This is problematic when it comes to their reconstruction through Structure from Motion (SfM). We argue in this paper that a more expressive and general assumption can be made around compressible 3D structures. The vision community, however, has hitherto struggled to formulate effective strategies for recovering such structures after projection without the aid of additional priors (e.g. temporal ordering, rigid substructures, etc.). In this paper we present a "prior-less" approach to solve compressible SfM. Specifically, we demonstrate how the problem of SfM - assuming compressible 3D structures - can be theoretically characterized as a block sparse dictionary learning problem. We validate our approach experimentally by demonstrating reconstructions of 3D structures that are intractable using current state-of-theart low-rank SfM approaches.
{"title":"Prior-Less Compressible Structure from Motion","authors":"Chen Kong, S. Lucey","doi":"10.1109/CVPR.2016.447","DOIUrl":"https://doi.org/10.1109/CVPR.2016.447","url":null,"abstract":"Many non-rigid 3D structures are not modelled well through a low-rank subspace assumption. This is problematic when it comes to their reconstruction through Structure from Motion (SfM). We argue in this paper that a more expressive and general assumption can be made around compressible 3D structures. The vision community, however, has hitherto struggled to formulate effective strategies for recovering such structures after projection without the aid of additional priors (e.g. temporal ordering, rigid substructures, etc.). In this paper we present a \"prior-less\" approach to solve compressible SfM. Specifically, we demonstrate how the problem of SfM - assuming compressible 3D structures - can be theoretically characterized as a block sparse dictionary learning problem. We validate our approach experimentally by demonstrating reconstructions of 3D structures that are intractable using current state-of-theart low-rank SfM approaches.","PeriodicalId":6515,"journal":{"name":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"55 7 1","pages":"4123-4131"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85823632","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}
Mingkui Tan, Shijie Xiao, Junbin Gao, Dong Xu, A. Hengel, Javen Qinfeng Shi
Trace-norm regularization plays an important role in many areas such as computer vision and machine learning. When solving general large-scale trace-norm regularized problems, existing methods may be computationally expensive due to many high-dimensional truncated singular value decompositions (SVDs) or the unawareness of matrix ranks. In this paper, we propose a proximal Riemannian pursuit (PRP) paradigm which addresses a sequence of trace-norm regularized subproblems defined on nonlinear matrix varieties. To address the subproblem, we extend the proximal gradient method on vector space to nonlinear matrix varieties, in which the SVDs of intermediate solutions are maintained by cheap low-rank QR decompositions, therefore making the proposed method more scalable. Empirical studies on several tasks, such as matrix completion and low-rank representation based subspace clustering, demonstrate the competitive performance of the proposed paradigms over existing methods.
{"title":"Proximal Riemannian Pursuit for Large-Scale Trace-Norm Minimization","authors":"Mingkui Tan, Shijie Xiao, Junbin Gao, Dong Xu, A. Hengel, Javen Qinfeng Shi","doi":"10.1109/CVPR.2016.633","DOIUrl":"https://doi.org/10.1109/CVPR.2016.633","url":null,"abstract":"Trace-norm regularization plays an important role in many areas such as computer vision and machine learning. When solving general large-scale trace-norm regularized problems, existing methods may be computationally expensive due to many high-dimensional truncated singular value decompositions (SVDs) or the unawareness of matrix ranks. In this paper, we propose a proximal Riemannian pursuit (PRP) paradigm which addresses a sequence of trace-norm regularized subproblems defined on nonlinear matrix varieties. To address the subproblem, we extend the proximal gradient method on vector space to nonlinear matrix varieties, in which the SVDs of intermediate solutions are maintained by cheap low-rank QR decompositions, therefore making the proposed method more scalable. Empirical studies on several tasks, such as matrix completion and low-rank representation based subspace clustering, demonstrate the competitive performance of the proposed paradigms over existing methods.","PeriodicalId":6515,"journal":{"name":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"49 1","pages":"5877-5886"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86334233","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}
Our goal is to recognize material categories using images and geometry information. In many applications, such as construction management, coarse geometry information is available. We investigate how 3D geometry (surface normals, camera intrinsic and extrinsic parameters) can be used with 2D features (texture and color) to improve material classification. We introduce a new dataset, GeoMat, which is the first to provide both image and geometry data in the form of: (i) training and testing patches that were extracted at different scales and perspectives from real world examples of each material category, and (ii) a large scale construction site scene that includes 160 images and over 800,000 hand labeled 3D points. Our results show that using 2D and 3D features both jointly and independently to model materials improves classification accuracy across multiple scales and viewing directions for both material patches and images of a large scale construction site scene.
{"title":"Geometry-Informed Material Recognition","authors":"Joseph DeGol, M. G. Fard, Derek Hoiem","doi":"10.1109/CVPR.2016.172","DOIUrl":"https://doi.org/10.1109/CVPR.2016.172","url":null,"abstract":"Our goal is to recognize material categories using images and geometry information. In many applications, such as construction management, coarse geometry information is available. We investigate how 3D geometry (surface normals, camera intrinsic and extrinsic parameters) can be used with 2D features (texture and color) to improve material classification. We introduce a new dataset, GeoMat, which is the first to provide both image and geometry data in the form of: (i) training and testing patches that were extracted at different scales and perspectives from real world examples of each material category, and (ii) a large scale construction site scene that includes 160 images and over 800,000 hand labeled 3D points. Our results show that using 2D and 3D features both jointly and independently to model materials improves classification accuracy across multiple scales and viewing directions for both material patches and images of a large scale construction site scene.","PeriodicalId":6515,"journal":{"name":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"68 1","pages":"1554-1562"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77725630","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}
Shenlong Wang, S. Fanello, Christoph Rhemann, S. Izadi, Pushmeet Kohli
This paper proposes a novel extremely efficient, fully-parallelizable, task-specific algorithm for the computation of global point-wise correspondences in images and videos. Our algorithm, the Global Patch Collider, is based on detecting unique collisions between image points using a collection of learned tree structures that act as conditional hash functions. In contrast to conventional approaches that rely on pairwise distance computation, our algorithm isolates distinctive pixel pairs that hit the same leaf during traversal through multiple learned tree structures. The split functions stored at the intermediate nodes of the trees are trained to ensure that only visually similar patches or their geometric or photometric transformed versions fall into the same leaf node. The matching process involves passing all pixel positions in the images under analysis through the tree structures. We then compute matches by isolating points that uniquely collide with each other ie. fell in the same empty leaf in multiple trees. Our algorithm is linear in the number of pixels but can be made constant time on a parallel computation architecture as the tree traversal for individual image points is decoupled. We demonstrate the efficacy of our method by using it to perform optical flow matching and stereo matching on some challenging benchmarks. Experimental results show that not only is our method extremely computationally efficient, but it is also able to match or outperform state of the art methods that are much more complex.
{"title":"The Global Patch Collider","authors":"Shenlong Wang, S. Fanello, Christoph Rhemann, S. Izadi, Pushmeet Kohli","doi":"10.1109/CVPR.2016.21","DOIUrl":"https://doi.org/10.1109/CVPR.2016.21","url":null,"abstract":"This paper proposes a novel extremely efficient, fully-parallelizable, task-specific algorithm for the computation of global point-wise correspondences in images and videos. Our algorithm, the Global Patch Collider, is based on detecting unique collisions between image points using a collection of learned tree structures that act as conditional hash functions. In contrast to conventional approaches that rely on pairwise distance computation, our algorithm isolates distinctive pixel pairs that hit the same leaf during traversal through multiple learned tree structures. The split functions stored at the intermediate nodes of the trees are trained to ensure that only visually similar patches or their geometric or photometric transformed versions fall into the same leaf node. The matching process involves passing all pixel positions in the images under analysis through the tree structures. We then compute matches by isolating points that uniquely collide with each other ie. fell in the same empty leaf in multiple trees. Our algorithm is linear in the number of pixels but can be made constant time on a parallel computation architecture as the tree traversal for individual image points is decoupled. We demonstrate the efficacy of our method by using it to perform optical flow matching and stereo matching on some challenging benchmarks. Experimental results show that not only is our method extremely computationally efficient, but it is also able to match or outperform state of the art methods that are much more complex.","PeriodicalId":6515,"journal":{"name":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"1 1","pages":"127-135"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82143987","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}
Seonghyeon Nam, Youngbae Hwang, Y. Matsushita, Seon Joo Kim
Modelling and analyzing noise in images is a fundamental task in many computer vision systems. Traditionally, noise has been modelled per color channel assuming that the color channels are independent. Although the color channels can be considered as mutually independent in camera RAW images, signals from different color channels get mixed during the imaging process inside the camera due to gamut mapping, tone-mapping, and compression. We show the influence of the in-camera imaging pipeline on noise and propose a new noise model in the 3D RGB space to accounts for the color channel mix-ups. A data-driven approach for determining the parameters of the new noise model is introduced as well as its application to image denoising. The experiments show that our noise model represents the noise in regular JPEG images more accurately compared to the previous models and is advantageous in image denoising.
{"title":"A Holistic Approach to Cross-Channel Image Noise Modeling and Its Application to Image Denoising","authors":"Seonghyeon Nam, Youngbae Hwang, Y. Matsushita, Seon Joo Kim","doi":"10.1109/CVPR.2016.186","DOIUrl":"https://doi.org/10.1109/CVPR.2016.186","url":null,"abstract":"Modelling and analyzing noise in images is a fundamental task in many computer vision systems. Traditionally, noise has been modelled per color channel assuming that the color channels are independent. Although the color channels can be considered as mutually independent in camera RAW images, signals from different color channels get mixed during the imaging process inside the camera due to gamut mapping, tone-mapping, and compression. We show the influence of the in-camera imaging pipeline on noise and propose a new noise model in the 3D RGB space to accounts for the color channel mix-ups. A data-driven approach for determining the parameters of the new noise model is introduced as well as its application to image denoising. The experiments show that our noise model represents the noise in regular JPEG images more accurately compared to the previous models and is advantageous in image denoising.","PeriodicalId":6515,"journal":{"name":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"1 1","pages":"1683-1691"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81715121","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}
RGBD scene recognition has attracted increasingly attention due to the rapid development of depth sensors and their wide application scenarios. While many research has been conducted, most work used hand-crafted features which are difficult to capture high-level semantic structures. Recently, the feature extracted from deep convolutional neural network has produced state-of-the-art results for various computer vision tasks, which inspire researchers to explore incorporating CNN learned features for RGBD scene understanding. On the other hand, most existing work combines rgb and depth features without adequately exploiting the consistency and complementary information between them. Inspired by some recent work on RGBD object recognition using multi-modal feature fusion, we introduce a novel discriminative multi-modal fusion framework for rgbd scene recognition for the first time which simultaneously considers the inter-and intra-modality correlation for all samples and meanwhile regularizing the learned features to be discriminative and compact. The results from the multimodal layer can be back-propagated to the lower CNN layers, hence the parameters of the CNN layers and multimodal layers are updated iteratively until convergence. Experiments on the recently proposed large scale SUN RGB-D datasets show that our method achieved the state-of-the-art without any image segmentation.
{"title":"Discriminative Multi-modal Feature Fusion for RGBD Indoor Scene Recognition","authors":"Hongyuan Zhu, Jean-Baptiste Weibel, Shijian Lu","doi":"10.1109/CVPR.2016.324","DOIUrl":"https://doi.org/10.1109/CVPR.2016.324","url":null,"abstract":"RGBD scene recognition has attracted increasingly attention due to the rapid development of depth sensors and their wide application scenarios. While many research has been conducted, most work used hand-crafted features which are difficult to capture high-level semantic structures. Recently, the feature extracted from deep convolutional neural network has produced state-of-the-art results for various computer vision tasks, which inspire researchers to explore incorporating CNN learned features for RGBD scene understanding. On the other hand, most existing work combines rgb and depth features without adequately exploiting the consistency and complementary information between them. Inspired by some recent work on RGBD object recognition using multi-modal feature fusion, we introduce a novel discriminative multi-modal fusion framework for rgbd scene recognition for the first time which simultaneously considers the inter-and intra-modality correlation for all samples and meanwhile regularizing the learned features to be discriminative and compact. The results from the multimodal layer can be back-propagated to the lower CNN layers, hence the parameters of the CNN layers and multimodal layers are updated iteratively until convergence. Experiments on the recently proposed large scale SUN RGB-D datasets show that our method achieved the state-of-the-art without any image segmentation.","PeriodicalId":6515,"journal":{"name":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","volume":"18 1","pages":"2969-2976"},"PeriodicalIF":0.0,"publicationDate":"2016-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90067423","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：