AI Paper Atlas

Research map / Detection and segmentation / CNN features

Real-Time Instance Segmentation With Discriminative Orientation Maps

Wentao Du, Zhiyu Xiang, Shuya Chen, Chengyu Qiao, Yiman Chen, Tingming Bai

ICCV 2021 · Detection and segmentation › CNN features · approach: Boundary-aware instance segmentation · 29 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
replace grid centers with box centroids as base positions (ablation)OrienMask · COCO val2017mask AP: 32.5 -> 33.3Improved
use larger anchors inherited from YOLOv4 (ablation)OrienMask · COCO val2017mask AP: 33.3 -> 33.8Improved
merge multi-scale pyramid features for P2 (fpn plus) (ablation)OrienMask · COCO val2017mask AP: 33.8 -> 34.1Improved
expand valid training area ratio from 1.0 to 1.2 (ablation)OrienMask · COCO val2017mask AP: 32.0 -> 32.5Improved
expand valid training area ratio from 1.2 to 1.6 (ablation)OrienMask · COCO val2017mask AP: 32.5 -> 31.9Hurt
increase orientation loss weight lambda from 5 to 20 (ablation)OrienMask · COCO val2017mask AP: 31.0 -> 32.5Improved
increase orientation loss weight lambda (ablation)OrienMask · COCO val2017box AP: 36.7 -> 37.4No clear effect
contraction ratio tau 0.6 instead of 0.4 (ablation)OrienMask · COCO val2017mask AP: 30.4 -> 32.5Improved
contraction ratio tau 0.6 instead of 0.8 (ablation)OrienMask · COCO val2017mask AP: 30.1 -> 32.5Improved
use OrienMask instead of YOLACTreal-time instance segmentation · COCO val2017mask AP: 28.9 -> 34.5Improved
use OrienMask instead of YOLACTreal-time instance segmentation · COCO val2017FPS: 45.9 -> 41.9Hurt
use OrienMask instead of BlendMask for mask constructionreal-time instance segmentation · COCO val2017mask feature memory (MB): 18.44M -> 1.27MImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 4 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