AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Learning Transformations To Reduce the Geometric Shift in Object Detection

Vidit Vidit, Martin Engilberge, Mathieu Salzmann

CVPR 2023 · Detection and segmentation › Open-set and few-shot detection · approach: Domain-adaptive object detection · 4 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
Learn homographies and aggregate their transformed features with target-domain self-training instead of using MT aloneFasterRCNN ResNet50 · Cityscapes → KITTI FoV adaptationCar AP@0.5: 78.3 -> 80.4Improved
Use learned transformations instead of PIT correctionOurs vs. MT+PIT · Cityscapes → KITTI FoV adaptationCar AP@0.5: 79.7 -> 80.4Improved
Use Mean Teacher pseudo-label adaptation rather than source-only trainingFasterRCNN ResNet50 · Cityscapes → KITTI FoV adaptationCar AP@0.5: 76.1 -> 78.3Improved
Optimize homographies instead of using FR+PIT (ablation)Ours-h · FoVx 70°Car AP@0.5: 91.4 -> 93.1Improved
Optimize homographies instead of using FR+PIT (ablation)Ours-h · FoVx 80°Car AP@0.5: 89.2 -> 91.8Improved
Optimize homographies instead of using FR+PIT (ablation)Ours-h · FoVx 90°Car AP@0.5: 85.9 -> 88.8Improved
Optimize homographies instead of using FR (ablation)Ours-h · FoVx 50°Car AP@0.5: 94.3 -> 94.1Hurt
Optimize only homography parameters, keeping the base and aggregator frozen (ablation)FasterRCNN ResNet50 · Cityscapes → KITTI FoV adaptationAP@0.5: FR 76.1; Ours-h 78.2Improved
Optimize only homography parameters, keeping the base and aggregator frozen (ablation)FasterRCNN ResNet50 · Cityscapes → MOT viewpoint adaptationAP@0.5: 49.8; FR value not reported in the supplied textImproved
Initialize all homographies to identity and learn a diverse set (ablation)Ours · Cityscapes → KITTI FoV adaptationAP@0.5: 79.5 AP@0.5Improved
Increase the number of learned homographies (ablation)Ours · Cityscapes → KITTI FoV adaptationAPImproved

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