AI Paper Atlas

Research map / Detection and segmentation / CNN features

Rethinking Detection Calibration: A Coordinate Perspective

Juyong Lee, Seungjin Jung, Jungmin Lee, Sunju Lee, Jongwon Choi

ECCV 2026 · Detection and segmentation › CNN features · approach: Box regression and loss design

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 coordinate-wise confidence re-encoding (ReDC) versus box-level IR calibrationDeformable-DETR · COCO minitest (in-domain)Cx1-ECE: 12.0 -> 7.7Improved
ReDC versus box-level calibrators on box-level calibrationDeformable-DETR · COCO minitest (in-domain)LaECE0: 7.8 -> 7.9No clear effect
ReDC versus IR for LaECE0 on CityscapesDeformable-DETR · Cityscapes minitest (in-domain)Cx1-ECE: 12.2 -> 6.9Improved
ReDC under domain shift to COCO-CDeformable-DETR · COCO-C minitest (domain shift)Cx1-ECE: 16.1 -> 11.9Improved
ReDC under domain shift to Foggy CityscapesDeformable-DETR · Foggy Cityscapes minitest (domain shift)Cx1-ECE: 13.7 -> 10.2Improved
add directional displacement estimation (ablation)ReDC · COCO (ablation, 5 seeds)LaECE0: 9.96 -> 9.88Improved
add directional displacement estimation (ablation)ReDC · COCO (ablation, 5 seeds)LaACE0: 23.14 -> 23.14No clear effect
add directional displacement estimation (ablation)ReDC · COCO (ablation, 5 seeds)Cx1-ECE: 11.22 -> 11.18No clear effect
add directional displacement estimation (ablation)ReDC · COCO (ablation, 5 seeds)Cy1-ECE: 11.90 -> 10.86Improved
ReDC versus probabilistic-detector calibration GP-NormalReDC · COCO coordinate-wise error correlationmean correlation: 0.0263 -> 0.1771Improved
apply ReDC to a different detector (DINO)DINO · COCO (model-agnostic)Cx1-ECE: 18.3 -> 9.4Improved
apply ReDC to Cascade R-CNNCascade R-CNN · COCO (model-agnostic)Da-CE: 19.3 -> 28.1Hurt

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