AI Paper Atlas

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

Adaptive Object Detection with Dual Multi-Label Prediction

Zhen Zhao, Yuhong Guo, Haifeng Shen, Jieping Ye

ECCV 2020 · Detection and segmentation › Open-set and few-shot detection · approach: Domain-adaptive object detection · 55 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 multi-label conditional adversary (MC) instead of no alignment (ablation)Faster R-CNN · Cityscapes to Foggy CityscapesmAP (%): 25.1 -> 36.6Improved
condition the adversary on multi-label predictions instead of unconditional adversary (ablation)MCAR · Cityscapes to Foggy CityscapesmAP (%): 33.7 -> 36.6Improved
add prediction consistency regularization (PR) (ablation)MCAR · Cityscapes to Foggy CityscapesmAP (%): 36.6 -> 38.8Improved
add prediction consistency regularization (PR) (ablation)MCAR with unconditional adversary · Cityscapes to Foggy CityscapesmAP (%): 31.0 -> 33.7Improved
add multi-label prediction loss (ablation)MCAR with unconditional adversary without PR · Cityscapes to Foggy CityscapesmAP (%): 30.5 -> 31.0Improved
add prediction regularization to MCMCAR (MC only) · PASCAL VOC to WatercolormAP (%): 54.4 -> 56.0Improved
add prediction regularization to MCMCAR (MC only) · PASCAL VOC to ComicmAP (%): 32.6 -> 33.5Improved
apply MCAR domain adaptation instead of source-onlySource-only Faster R-CNN · PASCAL VOC to WatercolormAP (%): 44.6 -> 56.0Improved
apply MCAR instead of source-onlySource-only Faster R-CNN · PASCAL VOC to ComicmAP (%): 19.7 -> 33.5Improved
use MCAR instead of SCLSCL · Cityscapes to Foggy CityscapesmAP (%): 37.9 -> 38.8Improved
reduce focal modulation gamma from 5 to 1 (ablation)MCAR adversary · PASCAL VOC to Watercolor (hyperparameter sensitivity)mAP (%): 54.4 -> 44.0Hurt
increase focal modulation gamma from 5 to 9 (ablation)MCAR adversary · PASCAL VOC to Watercolor (hyperparameter sensitivity)mAP (%): 54.4 -> 44.8Hurt

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