AI Paper Atlas

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

Enhancing Source-Free Domain Adaptive Object Detection with Low-confidence Pseudo Label Distillation

Ilhoon Yoon, Hyeongjun Kwon, Jin Kim, Junyoung Park, Hyunsung Jang, Kwanghoon Sohn

ECCV 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Domain-adaptive object detection · 29 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 low-confidence pseudo labels (LPL) to high-confidence pseudo labels (HPL) (ablation)Mean-Teacher SFOD · Cityscapes -> Foggy CityscapesmAP: 34.3 -> 38.9Improved
add feature-distance adaptive weights to LPLD loss (ablation)HPL+LPL · Cityscapes -> Foggy CityscapesmAP: 38.9 -> 40.4Improved
use KL-divergence for classification and drop regression loss in LPLD instead of cross-entropy plus regression (ablation)LPLD loss · Cityscapes -> Foggy CityscapesmAP: 36.6 -> 38.9 (without adaptive weights)Improved
add regression loss on top of cross-entropy for LPL (ablation)LPLD loss · Cityscapes -> Foggy CityscapesmAP: 37.2 -> 36.6Hurt
lower IoU threshold of LPL mining against HPL from 0.5 to 0.4 (ablation)LPL mining · Cityscapes -> Foggy CityscapesmAP: 38.3 -> 40.4Improved
apply background confidence threshold 0.99 to LPL mining instead of none (1.00) (ablation)LPL mining · Cityscapes -> Foggy CityscapesmAP: 39.3 -> 40.4Improved
use class-confidence threshold 0.9 for LPL instead of 0.4 (ablation)LPL mining · Cityscapes -> Foggy CityscapesmAP: 38.1 -> 40.4Improved
apply LPLD loss to Mean-TeacherMean-Teacher SFOD · Cityscapes -> Foggy Cityscapesfalse negatives: 11491 -> 2822 false negativesImproved
apply LPLD on Sim10k -> Cityscapes against PETSSFOD · Sim10k -> CityscapesAP50 of car: 57.8 (PETS) vs 49.4Hurt
apply LPLD on Pascal-VOC -> ClipartSFOD · Pascal-VOC -> ClipartmAP: 31.5 -> 34.0 (vs IRG)Improved

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