AI Paper Atlas

Research map / Detection and segmentation / CNN features

Instance and Panoptic Segmentation Using Conditional Convolutions

Zhi Tian, Bowen Zhang, Hao Chen, Chunhua Shen

TPAMI 2022 · Detection and segmentation › CNN features · approach: Boundary-aware instance segmentation · 80 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 relative coordinates to the mask head input (ablation)CondInst mask head · COCO val2017mask AP: 31.5 -> 35.6Improved
use absolute instead of relative coordinates (ablation)CondInst mask head · COCO val2017mask AP: 35.6 -> 32.0Hurt
add bottom-branch features to mask head input (ablation)CondInst mask head · COCO val2017mask AP: 31.3 -> 35.6Improved
increase mask head depth from 1 to 3 (ablation)CondInst mask head · COCO val2017mask AP: 30.5 -> 35.6Improved
increase mask head depth from 3 to 4 (ablation)CondInst mask head · COCO val2017mask AP: 35.6 -> 35.6No clear effect
increase mask head width from 2 to 8 (ablation)CondInst mask head · COCO val2017mask AP: 33.9 -> 35.6Improved
increase mask head width from 8 to 16 (ablation)CondInst mask head · COCO val2017mask AP: 35.6 -> 35.7No clear effect
vary bottom channels C_bottom from 2 to 16 (ablation)CondInst bottom branch · COCO val2017mask AP: 34.9 -> 35.6 -> 35.4No clear effect
use P2 instead of P3 for bottom module (ablation)CondInst · COCO val2017mask AP: 35.6 -> 36.0Improved
add upsampling of mask predictions (factor 2) (ablation)CondInst · COCO val2017mask AP: 34.6 -> 35.6Improved
use mask-based NMS instead of box NMS (ablation)CondInst · COCO val2017mask AP: 35.6 -> 35.6No clear effect
use panoptic-generated instance targets instead of instance supervision (ablation)CondInst panoptic · COCO panopticPQ: 42.94 -> 44.18Improved

Earlier methods this paper compares with

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