AI Paper Atlas

Research map / Video / Video segmentation and tracking

Synergistic Bleeding Region and Point Detection in Laparoscopic Surgical Videos

Jialun Pei, Zhangjun Zhou, Diandian Guo, Zhixi Li, Jing Qin, Bo Du, Pheng-Ann Heng

CVPR 2026 · Video › Video segmentation and tracking · approach: Sign language and physiological video analysis

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 edge generator (wavelet Laplacian edge prompts) (ablation)BlooDet · SurgBlood test setIoU: 61.20 -> 64.88Improved
add point memory modeling (ablation)BlooDet · SurgBlood test setPCK-5%: 42.83 -> 55.85Improved
add edge generator on top of point memory modeling for point detection (ablation)BlooDet · SurgBlood test setPCK-5%: 51.57 -> 55.85Improved
add edge generator alone (without point memory modeling) (ablation)BlooDet · SurgBlood test setPCK-5%: 45.14 -> 42.83Hurt
add wavelet Laplacian filter to the edge generator (ablation)BlooDet · SurgBlood test setIoU: 62.41 -> 64.88Improved
add high-resolution features F1 and F2 to the edge generator (ablation)BlooDet · SurgBlood test setIoU: 64.74 -> 64.88Improved
add optical flow for camera offset (background-averaged) (ablation)BlooDet · SurgBlood test setPCK-5%: 45.47 -> 55.85Improved
average optical flow over foreground instead of background (ablation)BlooDet · SurgBlood test setPCK-5%: 55.85 -> 41.02Hurt
average optical flow over whole frame instead of background (ablation)BlooDet · SurgBlood test setPCK-5%: 55.85 -> 49.59Hurt
add point prompt, mask map and mask memory cross-branch guidance (ablation)BlooDet · SurgBlood test setIoU, Dice, PCK-5%, PCK-10%Improved
dual-branch joint training versus task-specific modelsBlooDet · SurgBlood test setPCK-10%: 60.13 -> 83.69 vs D-CeLRImproved
add a point head to region-level models for dual-task learningSAM 2 · SurgBlood test setIoU: 63.51 -> 50.93Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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