AI Paper Atlas

Research map / Detection and segmentation / CNN features

YOLACT++ Better Real-Time Instance Segmentation

Daniel Bolya, Chong Zhou, Fanyi Xiao, Yong Jae Lee

TPAMI 2020 · Detection and segmentation › CNN features · approach: Boundary-aware instance segmentation · 581 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 fc-layer mask outputs with prototype and coefficient assembly (ablation)one-stage detector · COCO val2017mask AP: 20.7 -> 29.9Improved
use Fast NMS instead of standard NMS (ablation)YOLACT · COCO val2017mask AP and FPS: 30.0 -> 29.9 AP, 24.0 -> 33.5 FPSMixed
increase prototypes k from 8 to 32 (ablation)YOLACT · COCO val2017mask AP: 26.8 -> 27.7Improved
increase prototypes k from 32 to 256 (ablation)YOLACT · COCO val2017mask AP: 27.7 -> 27.7No clear effect
add semantic segmentation loss on P3 during training (ablation)YOLACT · COCOmask mAP: +0.4 mAPImproved
add 3 scales per FPN level to anchors (ablation)YOLACT · COCO val2017mask AP: 27.7 -> 30.2Improved
add 5 aspect ratios to anchors (ablation)YOLACT · COCO val2017mask AP: 27.7 -> 28.0Improved
add deformable convolution with interval 3 (ablation)YOLACT++ R-101 · COCO val2017mask AP: 31.7 -> 33.3Improved
use interval 3 instead of DCN in all blocks (ablation)YOLACT++ R-101 · COCO val2017mask AP: 33.5 -> 33.3No clear effect
add fast mask re-scoring (ablation)YOLACT++ R-101 · COCO val2017mask AP: 33.3 -> 34.4Improved
add fast mask re-scoring (ablation)YOLACT++ R-50 · COCO val2017mask AP: 32.7 -> 33.7Improved
replace Fast NMS with standard NMS in Mask R-CNN (ablation)Mask R-CNN · COCO val2017mask AP and FPS: 36.1 -> 35.8 AP, 8.6 -> 9.9 FPSMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 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