AI Paper Atlas

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

Human De-Occlusion: Invisible Perception and Recovery for Humans

Qiang Zhou, Shiyin Wang, Yitong Wang, Zilong Huang, Xinggang Wang

CVPR 2021 · Detection and segmentation › Open-set and few-shot detection · approach: Amodal and open-world instance segmentation · 33 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 the full two-stage human de-occlusion method instead of PCNetsAHP mask completion · AHP Syn. and AHP Realamodal mask ℓ1: 0.1959 -> 0.1500; 0.2218 -> 0.1635Improved
Use the full two-stage human de-occlusion method instead of PCNetsAHP mask completion · AHP Syn. and AHP RealIoU (amodal/invisible): 83.1/29.1 -> 84.6/43.7; 81.3/31.2 -> 86.1/40.3Improved
Use the full recovery method instead of PCNetsAHP content recovery · AHP Syn. and AHP RealFID: 18.50 -> 13.85; 28.30 -> 19.49Improved
Add a discriminator to mask completion (ablation)Mask completion baseline · AHP fixed validationamodal IoU: 84.0 -> 84.3Improved
Add modal segmentation (ablation)Mask completion baseline · AHP fixed validationamodal IoU: 84.3 -> 83.1Hurt
Add modal segmentation (ablation)Mask completion baseline · AHP fixed validationinvisible IoU: 30.3 -> 34.6Improved
Add parsing branches (ablation)Mask completion with discriminator and modal segmentation · AHP fixed validationinvisible IoU: 34.6 -> 40.0Improved
Add template pose masks (ablation)Mask completion with discriminator and modal segmentation · AHP fixed validationinvisible IoU: 34.6 -> 38.5Improved
Use background proportion w=0.3 instead of w=1.0 (ablation)Unet content recovery · AHP content recovery validationFID: 19.71 -> 17.85Improved
Add body-part attention (ablation)Parsing-guided content recovery · AHP content recovery validationFID: 19.66 -> 16.76Improved
Add visible-to-invisible context attention (ablation)Parsing-guided content recovery · AHP content recovery validationFID: 19.66 -> 17.37Improved
Fuse both PGA attention streams rather than cascading them (ablation)Parsing-guided content recovery · AHP content recovery validationFID: 14.66 -> 13.85Improved

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