Cross-Image Region Mining with Region Prototypical Network for Weakly Supervised Segmentation

2021-08-17 · Weide Liu, Xiangfei Kong, Tzu-Yi Hung, Guosheng Lin

Weakly supervised image segmentation trained with image-level labels usually suffers from inaccurate coverage of object areas during the generation of the pseudo groundtruth. This is because the object activation maps are trained with the classification objective and lack the ability to generalize. To improve the generality of the objective activation maps, we propose a region prototypical network RPNet to explore the cross-image object diversity of the training set. Similar object parts across images are identified via region feature comparison. Object confidence is propagated between regions to discover new object areas while background regions are suppressed. Experiments show that the proposed method generates more complete and accurate pseudo object masks, while achieving state-of-the-art performance on PASCAL VOC 2012 and MS COCO. In addition, we investigate the robustness of the proposed method on reduced training sets.

📄 PDF Abstract BibTeX arXiv:2108.07413

Code (1)

liuweide01/Cross-Image-Region-Mining-with-Region-Prototypical-Network-for-Weakly-Supervised-Segmentation 공식 구현 pytorch

Tasks

DiversityImage SegmentationObjectSemantic SegmentationWeakly supervised segmentation

Partition-A-Medical-Image: Extracting Multiple Representative Sub-regions for Few-shot Medical Image Segmentation

2023-09-20 · Yazhou Zhu, Shidong Wang, Tong Xin, Zheng Zhang 외

Few-shot Medical Image Segmentation (FSMIS) is a more promising solution for medical image segmentation tasks where high-quality annotations are naturally scarce. However, current mainstream methods primarily focus on ex…

Image Segmentation Medical Image Segmentation Semantic Segmentation

Learning Non-target Knowledge for Few-shot Semantic Segmentation

2022-05-10 · CVPR 2022 1 · Yuanwei Liu, Nian Liu, Qinglong Cao, Xiwen Yao 외

Existing studies in few-shot semantic segmentation only focus on mining the target object information, however, often are hard to tell ambiguous regions, especially in non-target regions, which include background (BG) an…

Contrastive Learning Few-Shot Semantic Segmentation Object Segmentation+1

Few-Shot Medical Image Segmentation via a Region-enhanced Prototypical Transformer

2023-09-09 · Yazhou Zhu, Shidong Wang, Tong Xin, Haofeng Zhang

Automated segmentation of large volumes of medical images is often plagued by the limited availability of fully annotated data and the diversity of organ surface properties resulting from the use of different acquisition…

Diversity Few-Shot Learning Image Segmentation Medical Image Segmentation+2

Texture based Prototypical Network for Few-Shot Semantic Segmentation of Forest Cover: Generalizing for Different Geographical Regions

2022-03-29 · Gokul P, Ujjwal Verma

Forest plays a vital role in reducing greenhouse gas emissions and mitigating climate change besides maintaining the world's biodiversity. The existing satellite-based forest monitoring system utilizes supervised learnin…

Few-Shot Semantic Segmentation Segmentation Semantic Segmentation

SLPD: Slide-level Prototypical Distillation for WSIs

2023-07-20 · Zhimiao Yu, Tiancheng Lin, Yi Xu

Improving the feature representation ability is the foundation of many whole slide pathological image (WSIs) tasks. Recent works have achieved great success in pathological-specific self-supervised learning (SSL). Howeve…

Representation Learning Self-Supervised Learning