Filters








3,310 Hits in 3.0 sec

Learning superpixel relations for supervised image segmentation

Marco Manfredi, Costantino Grana, Rita Cucchiara
2014 2014 IEEE International Conference on Image Processing (ICIP)  
In this paper we propose to extend the well known graph cut segmentation framework by learning superpixel relations and use them to weight superpixel-to-superpixel edges in a superpixel graph.  ...  Several superpixel-pair features are investigated and exploited to build a non-linear SVM to learn object boundary appearance.  ...  The main contribution of the paper is the formulation of a learning scheme able to model the relations between superpixels, applied to supervised binary object segmentation.  ... 
doi:10.1109/icip.2014.7025900 dblp:conf/icip/ManfrediGC14 fatcat:zqpjg4kuffdrpoxvb3gkxvmxde

Semi-supervised Learning for Large Scale Image Cosegmentation

Zhengxiang Wang, Rujie Liu
2013 2013 IEEE International Conference on Computer Vision  
This paper introduces to use semi-supervised learning for large scale image cosegmentation.  ...  between different images and the lack of segmentation groundtruth for guidance in cosegmentation.  ...  For supervised learning, each image is segmented individually with training images only, without considering the similarity of the common object shared between unsegmented images.  ... 
doi:10.1109/iccv.2013.56 dblp:conf/iccv/WangL13 fatcat:2itcdq5i2jhk3bf3fgxdxxefmu

Weakly supervised semantic segmentation with a multi-image model

Alexander Vezhnevets, Vittorio Ferrari, Joachim M. Buhmann
2011 2011 International Conference on Computer Vision  
For segmenting test images, we integrate them into MIM by means of a learned multiple kernel image similarity.  ...  The task of weakly supervised learning is to recover the latent labels y j i and to learn a classifier that predicts labels for each superpixel in a new image.  ...  Note, how we are able to recognize two very hard classes (boat and bird) substantially better than any previous methods (either weakly or fully supervised).  ... 
doi:10.1109/iccv.2011.6126299 dblp:conf/iccv/VezhnevetsFB11 fatcat:4l3tkxijl5hptbbswzmfr543mm

Learning from Pixel-Level Label Noise: A New Perspective for Semi-Supervised Semantic Segmentation [article]

Rumeng Yi, Yaping Huang, Qingji Guan, Mengyang Pu, Runsheng Zhang
2021 arXiv   pre-print
Then, we adopt a superpixel-based graph to represent the relations of spatial adjacency and semantic similarity between pixels in one image.  ...  This paper addresses semi-supervised semantic segmentation by exploiting a small set of images with pixel-level annotations (strong supervisions) and a large set of images with only image-level annotations  ...  Learning from Pixel-Level Label Noise: A New Perspective for Semi-Supervised Semantic Segmentation Rumeng Yi, Yaping Huang, Qingji Guan, Mengyang Pu, and Runsheng Zhang Abstract-This paper addresses semi-supervised  ... 
arXiv:2103.14242v1 fatcat:kwhr3d5kgjbrlbvvnhrkd4jdsm

Learning from Weak and Noisy Labels for Semantic Segmentation

Zhiwu Lu, Zhenyong Fu, Tao Xiang, Peng Han, Liwei Wang, Xin Gao
2017 IEEE Transactions on Pattern Analysis and Machine Intelligence  
This paper presents a weakly supervised sparse learning approach to the problem of noisily tagged image parsing, or segmenting all the objects within a noisily tagged image and identifying their categories  ...  By oversegmenting all the images into regions, we formulate noisily tagged image parsing as a weakly supervised sparse learning problem over all the regions, where the initial labels of each region are  ...  After that, a superpixel appearance model is learned for two purposes: to iteratively refine the semantic segmentation, and to predict superpixel labels of (i.e. to segment) an unseen test image.  ... 
doi:10.1109/tpami.2016.2552172 pmid:28113885 fatcat:gil3elyzy5fgnm3qy3c6fq7e44

Weakly-Supervised Image Annotation and Segmentation with Objects and Attributes

Zhiyuan Shi, Yongxin Yang, Timothy M. Hospedales, Tao Xiang
2017 IEEE Transactions on Pattern Analysis and Machine Intelligence  
correlations within and across superpixels.  ...  models on a variety of tasks including semantic segmentation, automatic image annotation and retrieval based on object-attribute associations.  ...  RELATED WORK Our work is related to a wide range of computer vision problems including image classification, object recognition, attribute learning, and semantic segmentation.  ... 
doi:10.1109/tpami.2016.2645157 pmid:28026753 fatcat:tdvm3fyzevevpkb52jlbwfc5f4

Weakly Supervised Image Annotation and Segmentation with Objects and Attributes [article]

Zhiyuan Shi, Yongxin Yang, Timothy M. Hospedales, Tao Xiang
2017 arXiv   pre-print
correlations within and across superpixels.  ...  models on a variety of tasks including semantic segmentation, automatic image annotation and retrieval based on object-attribute associations.  ...  RELATED WORK Our work is related to a wide range of computer vision problems including image classification, object recognition, attribute learning, and semantic segmentation.  ... 
arXiv:1708.02459v1 fatcat:amftoyrswvcw5latexbs3dupu4

Foreground Clustering for Joint Segmentation and Localization in Videos and Images [article]

Abhishek Sharma
2018 arXiv   pre-print
Exploiting the geometric relation between the superpixels and bounding boxes enables the transfer of segmentation cues to improve localization output and vice-versa.  ...  This paper presents a novel framework in which video/image segmentation and localization are cast into a single optimization problem that integrates information from low level appearance cues with that  ...  Note that for every bounding box i, x i ( superpixel indexing at bounding box level) and y (indexing at image level) are related by an indicator projection matrix P i of dimensions |S i | × n such that  ... 
arXiv:1811.10121v1 fatcat:bkxltqjygre3zm6fxu2ftap4rm

Weakly supervised semantic segmentation for social images

Wei Zhang, Sheng Zeng, Dequan Wang, Xiangyang Xue
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
In this paper, we learn a weakly supervised semantic segmentation model from social images whose labels are not pixellevel but image-level; furthermore, these labels might be noisy.  ...  Image semantic segmentation is the task of partitioning image into several regions based on semantic concepts.  ...  Related Works In the past years, image semantic segmentation has attracted a lot of attentions. Most of the existing works model the task as a fully supervised problem [32] . Shotton et al.  ... 
doi:10.1109/cvpr.2015.7298888 dblp:conf/cvpr/ZhangZWX15 fatcat:zjwopqvpv5e3jgeh5thfim7cmm

Self-Supervision with Superpixels: Training Few-shot Medical Image Segmentation without Annotation [article]

Cheng Ouyang, Carlo Biffi, Chen Chen, Turkay Kart, Huaqi Qiu, Daniel Rueckert
2020 arXiv   pre-print
imbalance problem in medical image segmentation; (3) We demonstrate the general applicability of the proposed approach for medical images using three different tasks: abdominal organ segmentation for  ...  To address this problem we make several contributions: (1) A novel self-supervised FSS framework for medical images in order to eliminate the requirement for annotations during training.  ...  This work is also supported by the UK Research and Innovation London Medical Imaging and Artificial Intelligence Centre for Value Based Healthcare.  ... 
arXiv:2007.09886v2 fatcat:tnmhky4sn5cv5ojlvglmgvcti4

Weakly Supervised Semantic Segmentation Using Superpixel Pooling Network

Suha Kwak, Seunghoon Hong, Bohyung Han
2017 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
To this end, we propose Superpixel Pooling Network (SPN), which utilizes superpixel segmentation of input image as a pooling layout to reflect low-level image structure for learning and inferring semantic  ...  We propose a weakly supervised semantic segmentation algorithm based on deep neural networks, which relies on image-level class labels only.  ...  , Global Frontier R&D Program on Human-Centered Interaction for Coexistence), which are funded by the Korean government (MSIP), NSF CAREER grant IIS-1453651, and ONR grant N00014-13-1-0762.  ... 
doi:10.1609/aaai.v31i1.11213 fatcat:q3jfr4ga5zhlvfai6skbt6reze

Probabilistic Graphlet Cut: Exploiting Spatial Structure Cue for Weakly Supervised Image Segmentation

Luming Zhang, Mingli Song, Zicheng Liu, Xiao Liu, Jiajun Bu, Chun Chen
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
In this paper, we present a new weakly supervised image segmentation algorithm by learning the distribution of spatially structured superpixel sets from image-level labels.  ...  Weakly supervised image segmentation is a challenging problem in computer vision field.  ...  Related Work Recently, several weakly supervised image segmentation methods have been proposed, which focus on developing statistical models to transfer image-level labels into superpixels unary or pairwise  ... 
doi:10.1109/cvpr.2013.249 dblp:conf/cvpr/ZhangSLLBC13 fatcat:s7avezxvjnc5lie2lxncgwfncu

Simultaneous semantic segmentation of a set of partially labeled images

Qiongjie Tian, Baoxin Li
2016 2016 IEEE Winter Conference on Applications of Computer Vision (WACV)  
In practice, there are voluminous on-line images, which unfortunately often have only incomplete image-level labels (tags) but would otherwise be potentially useful for a learning-based algorithm.  ...  Only limited efforts have been attempted on using such coarsely and incompletely labelled data for semantic segmentation.  ...  Related Works We briefly review below two classes of related research on semantic segmentation: those relying on fullysupervised learning and those utilizing only weaklysupervised learning.  ... 
doi:10.1109/wacv.2016.7477639 dblp:conf/wacv/TianL16 fatcat:7v6xsrpxhfdsjkcorfn3kbh67q

Representative Discovery of Structure Cues for Weakly-Supervised Image Segmentation

Luming Zhang, Yue Gao, Yingjie Xia, Ke Lu, Jialie Shen, Rongrong Ji
2014 IEEE transactions on multimedia  
This paper presents a weakly-supervised image segmentation algorithm that learns the distribution of spatially structural superpixel sets from image-level labels.  ...  Weakly-supervised image segmentation is a challenging problem with multidisciplinary applications in multimedia content analysis and beyond.  ...  RELATED WORK Recently, several weakly-supervised image segmentation methods [20] - [24] have been proposed, focusing on developing statistical models to transfer image-level labels into superpixels  ... 
doi:10.1109/tmm.2013.2293424 fatcat:m27nlglx5zfr5cff7ucpjkg52e

Weakly Supervised Deep Semantic Segmentation Using CNN and ELM with Semantic Candidate Regions

Xinying Xu, Guiqing Li, Gang Xie, Jinchang Ren, Xinlin Xie
2019 Complexity  
For fully supervised semantic segmentation, the task is achieved by a segmentation model trained using pixel-level annotations.  ...  Specifically, after segmenting each image into a set of superpixels, superpixels are automatically combined to achieve segmentation of candidate region according to the number of image-level labels.  ...  Acknowledgments The authors would like to express their gratitude for the support from the National Natural Science  ... 
doi:10.1155/2019/9180391 fatcat:65uncqoqrnesrh3dxyutjdbumi
« Previous Showing results 1 — 15 out of 3,310 results