AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

On the Effectiveness of Image Rotation for Open Set Domain Adaptation

Silvia Bucci, Mohammad Reza Loghmani, Tatiana Tommasi

ECCV 2020 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Source-free domain adaptation · 165 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
use rotation-recognition known/unknown separation instead of multi-binary classifierROS Stage I · Office-31AUC-ROC: 79.9 -> 91.5Improved
add center loss (ablation)ROS Stage I · Office-31AUC-ROC: 89.9 -> 91.5Improved
add anchor image (relative rotation) (ablation)ROS Stage I · Office-31AUC-ROC: 90.4 -> 91.5Improved
add rotation score to normality score (ablation)ROS Stage I · Office-31AUC-ROC: 89.6 -> 91.5Improved
add entropy score to normality score (ablation)ROS Stage I · Office-31AUC-ROC: 88.3 -> 91.5Improved
add center loss and anchor together (ablation)ROS Stage I · Office-31AUC-ROC: 87.1 -> 91.5Improved
use rotation recognition instead of GRL domain discriminator in Stage II (ablation)ROS Stage II · Office-31HOS: 85.1 -> 85.9Improved
use relative rotation with anchor in Stage II (ablation)ROS Stage II · Office-31HOS: 84.9 -> 85.9Improved
add entropy loss in Stage II (without anchor) (ablation)ROS Stage II · Office-31HOS: 85.0 vs 84.9No clear effect
apply rotation-based adaptation in Stage II (ablation)ROS · Office-31cosine distance source-target: 0.188 -> 0.109Improved
use HOS rather than OS for evaluationUAN · Office-HomeOS vs UNK: OS 72.5 vs UNK 0.0Mixed
increase openness (fewer known classes)STA and OSBP · Office-HomeHOSHurt

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