AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Unsupervised Hierarchical Semantic Segmentation With Multiview Cosegmentation and Clustering Transformers

Tsung-Wei Ke, Jyh-Jing Hwang, Yunhui Guo, Xudong Wang, Stella X. Yu

CVPR 2022 · Detection and segmentation › Open-set and few-shot detection · approach: Unsupervised object and part discovery · 34 citations (OpenAlex)

Check your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use multiview cosegmentation instead of single-view training (ablation)HSG · VOC valmIoU: 13.0 -> 40.9Improved
Add goodness-of-grouping loss (ablation)HSG with edge-based grouping loss · VOC valmIoU: 13.0 -> 13.8; 40.9 -> 41.7Improved
Add hierarchical feature-grouping contrastive loss (ablation)HSG with edge-based and goodness-of-grouping losses · VOC valmIoU: 13.8 -> 14.0; 41.7 -> 41.9Improved
Use the learned clustering transformer rather than KMeans clustering (ablation)HSG features · VOC valmIoU: 41.2 -> 41.9Improved
Use the learned clustering transformer rather than NCut clustering (ablation)HSG features · VOC valmIoU: 41.3 -> 41.9Improved
Use the learned clustering transformer rather than FINCH clustering (ablation)HSG features · VOC valmIoU: 40.6 -> 41.9Improved
Use hierarchical multiview feature grouping instead of image-wise instance discriminationUnsupervised semantic segmentation · Cityscapes valmIoU: 24.6 -> 32.5Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy