AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

CLCC: Contrastive Learning for Color Constancy

Yi-Chen Lo, Chia-Che Chang, Hsuan-Chao Chiu, Yu-Hao Huang, Chia-Ping Chen, Yu-Lin Chang, Kevin Jou

CVPR 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 56 citations (OpenAlex)

Check 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 CLCC contrastive learning with Full-Aug to SqueezeNet-FC4SqueezeNet-FC4 · NUS-8mean angular error: 2.23 -> 1.84Improved
Add CLCC contrastive learning with Full-Aug to SqueezeNet-FC4SqueezeNet-FC4 · NUS-8worst-25% angular error: 5.15 -> 4.20Improved
Add CLCC contrastive learning with Full-Aug to SqueezeNet-FC4SqueezeNet-FC4 · Gehlermean angular error: 1.65 -> 1.44Mixed
Add CLCC contrastive learning with Full-Aug to SqueezeNet-FC4SqueezeNet-FC4 · Gehlerworst-25% angular error: 3.78 -> 3.48Improved
Use neutral-color WB-Aug instead of no CLCC augmentation (ablation)SqueezeNet-FC4 · NUS-8mean angular error: 2.23 -> 1.93Improved
Use neutral-color WB-Aug instead of no CLCC augmentation (ablation)SqueezeNet-FC4 · NUS-8worst-25% angular error: 5.15 -> 4.30Improved
Use full-color mapping (Full-Aug) instead of neutral-color mapping (WB-Aug) (ablation)CLCC on SqueezeNet-FC4 · NUS-8mean angular error: 1.93 -> 1.84Improved
Use full-color mapping (Full-Aug) instead of neutral-color mapping (WB-Aug) (ablation)CLCC on SqueezeNet-FC4 · NUS-8worst-25% angular error: 4.30 -> 4.20Improved
Add CLCC contrastive learningSqueezeNet-FC4 · Gehler sparse illuminant cluster (16 data points)worst-case error: 3.24 -> 2.31Improved

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