AI Paper Atlas

Research map / Detection and segmentation / Weakly supervised detection and segmentation

BoxSnake: Polygonal Instance Segmentation with Box Supervision

Rui Yang, Lin Song, Yixiao Ge, Xiu Li

ICCV 2023 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised semantic segmentation · 41 citations (OpenAlex)

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 distance-aware pairwise losses on top of the point-based unary loss (ablation)BoxSnake · COCO val2017mask AP: 23.9 -> 31.1Improved
add local pairwise term (unary + local) (ablation)BoxSnake · COCO val2017mask AP: 23.9 -> 30.8Improved
add global pairwise term (unary + global) (ablation)BoxSnake · COCO val2017mask AP: 23.9 -> 29.8Improved
use CIoU/GIoU point-based unary loss instead of Dice on P3 (ablation)BoxSnake · COCO val2017mask AP: 19.3 -> 23.9Improved
add clipping strategy instead of scaling polygon to P2 (ablation)BoxSnake · COCO val2017mask AP: 29.8 -> 31.1Improved
increase pairwise-loss resolution from P3 to P2 size (ablation)BoxSnake · COCO val2017mask AP: 29.6 -> 29.8No clear effect
vary local pairwise window size and dilation (ablation)BoxSnake · COCO val2017mask AP: 30.8 / 30.9 / 31.1No clear effect
use ellipse instead of square initial polygon (ablation)BoxSnake · COCO val2017mask AP: 30.7 -> 31.1Improved
use proposed local pairwise loss instead of CRF loss (ablation)BoxSnake · COCO val2017mask AP: 30.2 -> 31.1Improved
set sigmoid temperature tau to 0.1 rather than 0.5 (ablation)BoxSnake · COCO val2017mask AP: 29.9 -> 31.1Improved
polygon output with box supervision instead of mask output (BoxInst)BoxSnake · Cityscapes valmask AP: 22.4 -> 26.3Improved
use BoxSnake polygon head instead of BoxInst mask headbox-supervised instance segmentation · COCO val2017 (R50, 1x)mask AP: 30.7 -> 31.1Improved

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