AI Paper Atlas

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

Rotation Consistent Margin Loss for Efficient Low-Bit Face Recognition

Yudong Wu, Yichao Wu, Ruihao Gong, Yuanhao Lv, Ken Chen, Ding Liang, Xiaolin Hu, Xianglong Liu, Junjie Yan

CVPR 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Margin-based face recognition losses · 48 citations (OpenAlex)

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 RCM loss to low-bit ResNet184-bit ResNet18 face recognition · MegaFace, 1M distractorsrank-1 identification accuracy: 87.74 -> 88.56Improved
Add RCM loss to low-bit ResNet183-bit ResNet18 face recognition · MegaFace, 1M distractorsrank-1 identification accuracy: 72.84 -> 74.34Improved
Add RCM loss to MobileNetV24-bit MobileNetV2 face recognition · MegaFace, 1M distractorsrank-1 identification accuracy: 77.90 -> 80.12Improved
Add RCM loss to MobileNetV23-bit MobileNetV2 face recognition · MegaFace, 1M distractorsrank-1 identification accuracy: 59.59 -> 62.58Improved
Add RCM loss instead of ArcFace+Mimic4-bit ResNet18 · LFWverification accuracy: 98.68 -> 98.91Improved
Add RCM loss instead of ArcFace+Mimic4-bit ResNet18 · YTFverification accuracy: 94.76 -> 94.95Improved
Add feature mimic supervision to ArcFace (ablation)4-bit ResNet18 · MegaFace, 1M distractorsrank-1 identification accuracy: 67.75 -> 69.04Improved
Replace ArcFace+Mimic with RCM loss (ablation)4-bit ResNet18 · MegaFace, 1M distractorsrank-1 identification accuracy: 69.04 -> 71.57Improved
Use RCM angular-margin formula LA rather than LB or LC (ablation)4-bit ResNet18 · MegaFaceaccuracyImproved
Set the RCM scaling parameter to λ = 5.0 rather than other tested values (ablation)4-bit ResNet18 with LA · MegaFaceaccuracyMixed
Add RCM loss to Dorefa-Net quantization (ablation)4-bit ResNet18 · MegaFace, 1M distractorsrank-1 identification accuracy: 90.59 -> 91.55Improved
Add RCM loss to DSQ quantization (ablation)4-bit ResNet18 · MegaFace, 1M distractorsrank-1 identification accuracy: 88.98 -> 89.46Improved

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