AI Paper Atlas

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

Learning and Aligning Click-Aware Shape Prior for Interactive Amodal Instance Segmentation

Junjie Chen, Junwei Lin, Ren Hong, Shengjie Liu, Yuming Fang, Feng Qian, Yifan Zuo

CVPR 2026 · Detection and segmentation › Open-set and few-shot detection · approach: Amodal and open-world instance segmentation

Read the paperCheck 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
add standard shape prior retrieval to a clicks-only model (ablation)ClickPriorNet · KINSmIoU occ (3 clicks): 54.02 -> 58.17Improved
replace standard shape prior with click-aware shape prior (triplet loss) (ablation)ClickPriorNet · KINSmIoU occ (3 clicks): 58.17 -> 67.86Improved
add deformable-attention prior alignment (ablation)ClickPriorNet · KINSmIoU occ (3 clicks): 67.86 -> 73.16Improved
add click-aware shape prior to a clicks-only model (NoC85) (ablation)ClickPriorNet · KINSNoC85 occ: 2.85 -> 2.40Improved
use ClickPriorNet instead of MFP+C2Finteractive amodal segmentation · KINSNoC90 occ: 4.65 -> 4.31Improved
combine MFP with C2F-Seg shape priorMFP · KINSNoC90 occ: 4.76 -> 4.65Improved
use ClickPriorNet instead of MFP+C2F with 1 clickinteractive amodal segmentation · KINS, 1 clickmIoU occ: 55.11 -> 56.32Improved
use ClickPriorNet instead of MFP+C2F with 5 clicksinteractive amodal segmentation · KINS, 5 clicksmIoU occ: 73.35 -> 77.72Improved
use ClickPriorNet instead of MFP+C2F with 3 clicksinteractive amodal segmentation · COCOA cls, 3 clicksmIoU occ: 41.94 -> 49.39Improved
use ClickPriorNet instead of MFP+C2F with 3 clicksinteractive amodal segmentation · KINS, 3 clicksmIoU occ: MFP+C2F 3 clicks 67.70 vs Ours 73.16Mixed
use too small top-N (N = 6) in triplet loss (ablation)ClickPriorNet · KINSNoC85 occ and mIoU occHurt
use too few retrieved priors (K = 4), or K above 10 (ablation)ClickPriorNet · KINSNoC85 occ and mIoU occMixed

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