检索结果-内蒙古大学图书馆

Joint Color-Spatial-directional Clustering and Region Merging (JCSd-RM) for Unsupervised rgb-d image segmentation

IEEE TRANSACTIONS ON PATTERN ANALYSIS ANd MACHINE INTELLIGENCE 2016年第11期38卷 2255-2268页

作者： Hasnat, Md. Abul Alata, Olivier Tremeau, Alain Univ St Etienne Lab Hubert Curien St Etienne France Univ St Etienne Lab Hubert Curien UMR CNRS 5516 Dept Comp Sci St Etienne France

Recent advances in depth imaging sensors provide easy access to the synchronized depth with color, called rgb-d image. In this paper, we propose an unsupervised method for indoor rgb-d image segmentation and analysis. We consider a statistical image generation model based on the color and geometry of the scene. Our method consists of a joint color-spatial-directional clustering method followed by a statistical planar region merging method. We evaluate our method on the NYU depth database and compare it with existing unsupervised rgb-d segmentation methods. Results show that, it is comparable with the state of the art methods and it needs less computation time. Moreover, it opens interesting perspectives to fuse color and geometry in an unsupervised manner.

关键词： Unsupervised clustering rgb-d image segmentation directional distributions Bregman divergence mixture model region adjacency graph region merging

来源：评论

学校读者我要写书评

暂无评论

Multi Scale CRF Based rgb-d image segmentation Using Inter Frames Potentials 2

Multi Scale CRF Based RGB-D Image Segmentation Using Inter F...

引用

2nd RSI/ISM International Conference on Robotics and Mechatronics (ICRoM)

作者： Hamedani, Taha Harati, Ahad Ferdowsi Univ Mashhad Robot Percept Lab Mashhad Iran

ISBN: (纸本)9781479967438

This paper proposed a novel multi-scale approach to solve energy minimization problem which can be used to deal with indoor scene labeling problem. The principal idea of all multi-scale algorithms is solving their finer problems to efficiently initialize coarser. We propose the use of both color and depth information which is captured by Microsoft Kinect sensor. In order to create our energy function, we use the Conditional Random Field (CRF) approach, and add our geometrical constraint to pairwise potential as regions extraction method based on both edge of rgb and Range image and definition of Cliques between two consecutive frames. We evaluate our method on challenging NYU v1 dataset and Experimental results show that our proposed method reached 2.35 for Hausdorff criterion and enhances the time of image segmentation.

关键词： Microsoft Kinect sensor rgb-d image segmentation Multi-scale CRF normal vector

来源：评论

学校读者我要写书评

暂无评论

Superpixel Based rgb-d image segmentation Using Markov Random Field

Superpixel Based RGB-D Image Segmentation Using Markov Rando...

引用

International Symposium on Artificial Intelligence and Signal Processing (AISP)

作者： Hamedani, Taha Zarei, Ramin Harati, Ahad Ferdowsi Univ Mashhad Robot Percept Lab Mashhad Iran

ISBN: (纸本)9781479988181

In this work we proposed a novel super pixel based segmentation approach to solve energy minimization problem which can be used to deal with indoor scene labeling problem. We used Range data beside color image captured from Kinect sensor. This sensor enables us to use 3d features of structure like normal vector and 2d color features. We extracted the region of scene as super pixel based on the both color and direction change;and, consequently, we constructed our graphical model on these regions and apply markov random field inference to assign efficient labels to them. Our evaluation on 30 scenes of challenging NYU v1 dataset shows that our proposed method reached higher values of "Correct detection" and lower rate of "Missed instances" and "Noise instances" criteria according to Hoover evaluation method.

关键词： Microsoft Kinect sensor rgb-d image segmentation Markov Random Field normal vector

来源：评论

学校读者我要写书评

暂无评论

GRAPH-BASEd rgb-d image segmentation USING COLOR-dIRECTIONAL-REGION MERGING 44

GRAPH-BASED RGB-D IMAGE SEGMENTATION USING COLOR-DIRECTIONAL...

引用

44th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Pan, Xiong Zhang, Zejun Liu, Yizhang Yang, Changcai Chen, Qiufeng Cheng, Li Lin, Jiaxiang Chen, Riqing Fujian Agr & Forestry Univ Coll Comp & Informat Sci Digital Fujian Inst Big Data Agr & Forestry Fuzhou Fujian Peoples R China Fujian Agr & Forestry Univ Jinshan Colledge Fuzhou Fujian Peoples R China

ISBN: (纸本)9781479981311

Color and depth information provided simultaneously in rgb-d images can be used to segment scenes into disjoint regions. In this paper, a graph-based segmentation method for rgb-d image is proposed, in which an adaptive data-driven combination of color- and normal-variation is presented to construct dissimilarity between two adjacent pixels and a novel region merging threshold exploiting normal information in adjacent regions is proposed to control the proceeding of the region merging. We evaluate our method on the NYU-v2 depth database and compare it with several published rgb-d partition methods. The experimental results show that our method is comparable with the state-of-the-art methods and provides more details of structures in the scene.

关键词： rgb-d image segmentation graph-based segmentation data-driven weight color-normal-region merging

来源：评论

学校读者我要写书评

暂无评论

rgb-d image segmentation BASEd ON MULTIPLE RANdOM WALKERS

RGB-D IMAGE SEGMENTATION BASED ON MULTIPLE RANDOM WALKERS

引用

IEEE International Conference on image Processing

作者： Se-Ho Lee Won-dong Jang Byung Kwan Park Chang-Su Kim School of Electrical Engineering Korea University Video Tech. Lab. SK Telecom

ISBN: (纸本)9781467399623

A novel rgb-d image segmentation algorithm is proposed in this work. This is the first attempt to achieve image segmentation based on the theory of multiple random walkers (MRW). We construct a multi-layer graph, whose nodes are superpixels divided with various parameters. Also, we set an edge weight to be proportional to the similarity of color and depth features between two adjacent nodes. Then, we segment an input rgb-d image by employing MRW simulation. Specifically, we decide the initial probability distribution of agents so that they are far from each other. We then execute the MRW process with the repulsive restarting rule, which makes the agents repel one another and occupy their own exclusive regions. Experimental results show that the proposed MRW image segmentation algorithm provides competitive segmentation performances, as compared with the conventional state-of-the-art algorithms.

关键词： Multiple random walkers segmentation Random walk rgb-d image segmentation Random walk Probability distribution image segmentation algorithm images Agents image segmentation adjacent nodes

来源：评论

学校读者我要写书评

暂无评论

Indoor Scene Understanding with rgb-d images: Bottom-up segmentation, Object detection and Semantic segmentation

引用

INTERNATIONAL JOURNAL OF COMPUTER VISION 2015年第2期112卷 133-149页

作者： Gupta, Saurabh Arbelaez, Pablo Girshick, Ross Malik, Jitendra Univ Calif Berkeley Berkeley CA 94720 USA Univ Los Andes Bogota Colombia

In this paper, we address the problems of contour detection, bottom-up grouping, object detection and semantic segmentation on rgb-d data. We focus on the challenging setting of cluttered indoor scenes, and evaluate our approach on the recently introduced NYU-depth V2 (NYUd2) dataset (Silberman et al., ECCV, 2012). We propose algorithms for object boundary detection and hierarchical segmentation that generalize the approach of Arbelaez et al. (TPAMI, 2011) by making effective use of depth information. We show that our system can label each contour with its type (depth, normal or albedo). We also propose a generic method for long-range amodal completion of surfaces and show its effectiveness in grouping. We train rgb-d object detectors by analyzing and computing histogram of oriented gradients on the depth image and using them with deformable part models (Felzenszwalb et al., TPAMI, 2010). We observe that this simple strategy for training object detectors significantly outperforms more complicated models in the literature. We then turn to the problem of semantic segmentation for which we propose an approach that classifies superpixels into the dominant object categories in the NYUd2 dataset. We design generic and class-specific features to encode the appearance and geometry of objects. We also show that additional features computed from rgb-d object detectors and scene classifiers further improves semantic segmentation accuracy. In all of these tasks, we report significant improvements over the state-of-the-art.

关键词： rgb-d contour detection rgb-d image segmentation rgb-d object detection rgb-d semantic segmentation rgb-d scene classification

来源：评论

学校读者我要写书评

暂无评论

Semantic segmentation of rgb-d images using 3dand Local Neighbouring Features

Semantic Segmentation of RGB-D Images using 3Dand Local Neig...

引用

International Conference on digital image Computing: Techniques and Applications

作者： Fooladgar, Fahimeh Kasaei, Shohreh Sharif Univ Technol Dept Comp Engn Tehran Iran

ISBN: (纸本)9781467367950

3d scene understanding is one of the most important problems in the field of computer vision. Although, in the past decades, considerable attention has been devoted on the 2d scene understanding problem, now with the development of the depth sensors (like Microsoft Kinect), the 3d scene understanding has become a very challenging task. Traditionally, the scene understanding problem was considered as the semantic labeling of each image pixel. Semantic labeling of rgb-d images has not attained a comparable success, as the rgb semantic labeling, due to the lack of a challenging dataset. With the introduction of an rgb-d dataset, called NYU-V2, it became possible to propose a novel method to improve the labeling accuracy. In this paper, a semantic segmentation algorithm for rgb-d images is presented. The concentration of the proposed algorithm is on the feature description and classification steps. In the feature description step, the more discriminative features from rgb images and the 3d point cloud data are grouped with local neighboring features to incorporate their context into the classification step. In the classification step, a pairwise multi-class conditional random field framework is utilized in which the unary potential function is considered as the probabilistic output of a random forest classifier. The proposed algorithm is evaluated on the NYU-V2 dataset and the performance is compared to that of other methods presented in the literature. The proposed algorithm achieves the state-of-the-art results on the NYU-V2 dataset.

关键词： rgb-d image segmentation 3d features semantic scene labeling

来源：评论

学校读者我要写书评

暂无评论

Enhance Support Relation Extraction Accuracy Using Improvement Of segmentation In rgb-d images 3

Enhance Support Relation Extraction Accuracy Using Improveme...

引用

3rd International Conference on Pattern Analysis and image Analysis (IPRIA)

作者： Ahmadi, Shokouh S. Khotanlou, Hassan Bu Ali Sina Univ RIV Lab Dept Comp Engn Hamadan Iran

ISBN: (纸本)9781509064540

Todays, increasing in machine vision fields and applications make it necessary to have accurate scene understanding and analyzing. Support relation extraction is one of the most important and critical problem in robotic and machine vision task. In this article, we enhance support relation extraction accuracy using improvement of segmentation. Having the depth, moreover the color, in rgb-d images enable us to obtain accurate and precise support relation. In this paper an approach is also presented to redress discontinuities in point cloud occurred while recording. Experimental result shows the accuracy of the extracted support relation will be significantly increase after segmentation improvement.

关键词： Support relation extraction 3d segmentation Scene understanding rgb-d image segmentation

来源：评论

学校读者我要写书评

暂无评论

Parsing Indoor Scenes from rgb-d image Using Superpixel and Region Merging 17

Parsing Indoor Scenes from RGB-D Image Using Superpixel and ...

引用

IEEE Int Conf on Parallel and distributed Processing with Applications, Big data and Cloud Computing, Sustainable Computing and Communications, Social Computing and Networking (ISPA/BdCloud/SocialCom/SustainCom)

作者： He, Kai Zhang, Zejun Pan, Xiong Yang, Changcai Yang, Sheng Lu, Xiaoliang Chen, Riqing Fujian Agr & Forestry Univ Coll Comp & Informat Sci Fuzhou 350002 Peoples R China Wuyi Univ Sch Math & Comp Sci Wuyishan 354300 Peoples R China MingByte Technol Co Ltd Qingdao 266000 Peoples R China Fujian Agr & Forestry Univ Sch Math & Informat Sci Fuzhou 350002 Peoples R China

ISBN: (纸本)9781728143286

This paper presents a new method of parsing indoor scene from an rgb-d image using superpixel of the rgb image and region merging of depth information. The goal of parsing indoor scene is to reconstruct the interior wall and floor scene based on an rgb-d image, the architecture of Manhattan structure. Firstly, an original rgb image of an indoor scene is segmented into superpixels using SLIC method, and a region merging procedure is used to iteratively merge adjacent superpixels possessing some identical properties in rgb- and depth-channel, which assures that the pixels in a merged region have basically identical normal, which is essential for reconstructing indoor wall and floor plane from a rgb-d image. Secondly, the obtained regions of the scene are translated into plane surface using normal information of regions, and then wall segments and walls are extracted to remove the influence of furniture and persons in the indoor scene on the reconstruction of indoor wall and floor plane. Finally, the Manhattan structure wall and floor scene of an indoor scene can be reconstructed by using dynamic programming method on candidate wall segments. The experiments show that the proposed method obtains better 3d structures than the ones the state-of-the-art produced.

关键词： Parsing indoor scene rgb-d image segmentation Superpixel region merging 3d reconstruction

来源：评论

学校读者我要写书评

暂无评论

Semantic segmentation of rgb-d images using 3d and Local Neighbouring Features

Semantic Segmentation of RGB-D Images using 3D and Local Nei...

引用

International Conference on digital image Computing: Techniques and Applications

作者： Fahimeh Fooladgar Shohreh Kasaei Department of Computer Engineering Sharif University of Technology

ISBN: (纸本)9781467367967

关键词： rgb-d image segmentation 3d features Semantic scene labeling

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：