AI Paper Atlas

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

A Robust Learning Approach to Domain Adaptive Object Detection

Mehran Khodabandeh, Arash Vahdat, Mani Ranjbar, William G. Macready

ICCV 2019 · Detection and segmentation › Open-set and few-shot detection · approach: Domain-adaptive object detection · 287 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
retrain on noisy target-domain pseudo boxes with robust label correctionFaster R-CNN · SIM 10K to CityscapesAP: 31.08 -> 42.56Improved
robust noisy-label training instead of plain pseudo-labelingPseudo-labeling · SIM 10K to CityscapesAP: 39.05 -> 42.56Improved
add bounding box refinement (Box-R) (ablation)Noisy Labeling · SIM 10K to CityscapesAP: 41.28 -> 41.83Improved
add false negative correction (FN-Cor) (ablation)Noisy Labeling · SIM 10K to CityscapesAP: 41.83 -> 42.56Improved
add Box-R (ablation)Noisy Labeling · Cityscapes to Foggy CityscapesmAP: 35.62 -> 36.06Improved
add FN-Cor (ablation)Noisy Labeling · Cityscapes to Foggy CityscapesmAP: 36.06 -> 36.45Improved
robust noisy-label training vs Feature LearningFaster R-CNN · Cityscapes to Foggy CityscapesmAP: 34.70 -> 36.45Improved
add Box-R (ablation)Noisy Labeling · KITTI to CityscapesAP: 42.03 -> 42.39Improved
add FN-Cor (ablation)Noisy Labeling · KITTI to CityscapesAP: 42.39 -> 42.98Improved
add Box-R (ablation)Noisy Labeling · Cityscapes to KITTIAP: 76.36 -> 76.93Improved
add FN-Cor (ablation)Noisy Labeling · Cityscapes to KITTIAP: 76.93 -> 77.61Improved
robust training vs Feature LearningFaster R-CNN · Cityscapes to KITTIAP: 73.76 -> 77.61Improved

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