AI Paper Atlas

Research map / Detection and segmentation / CNN features

RecursiveDet: End-to-End Region-Based Recursive Object Detection

Jing Zhao, Li Sun, Qingli Li

ICCV 2023 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 7 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
share decoder parameters across cascade stages (ablation)Sparse R-CNN · COCO val2017 (R50)AP: 42.8 -> 42.4 AP, 106M -> 40M paramsMixed
reuse dynamic layer and out layer within a stage (ablation)Sparse R-CNN with shared stages · COCO val2017 (R50)AP: 42.4 -> 43.6Improved
keep first decoder stage independent of recursive stages (ablation)RecursiveDet · COCO val2017 (R50)AP: 43.6 -> 43.9Improved
add bounding box positional encoding to the recursive decoder (ablation)RecursiveDet · COCO val2017 (R50)AP: 43.9 -> 44.4Improved
add bounding box PE to non-recursive Sparse R-CNN (ablation)Sparse R-CNN · COCO val2017 (R50)AP: 42.8 -> 43.1Improved
add centerness-based PE (ablation)RecursiveDet · COCO val2017 (R50)AP: 44.4 -> 45.2Improved
make centerness mask learnable instead of static (ablation)RecursiveDet · COCO val2017 (R50)AP: 45.2 -> 44.9Hurt
predict center offsets to adjust centerness peak (ablation)RecursiveDet · COCO val2017 (R50)AP: 45.2 -> 44.5Hurt
apply RecursiveDet to Sparse R-CNN with ResNet-50Sparse R-CNN · COCO val2017AP: 45.0 -> 46.5 AP, 106M -> 55M paramsImproved
apply RecursiveDet to AdaMixer with ResNet-50AdaMixer · COCO val2017AP: 47.0 -> 47.9 AP, 135M -> 43M paramsImproved
apply RecursiveDet to DiffusionDet with ResNet-101DiffusionDet · COCO val2017AP: 46.6 -> 46.9 AP, 130M -> 76M paramsImproved
add test-time augmentationRecursiveDet · COCO test-dev (Swin-B)AP: 53.1 -> 55.1Improved

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