AI Paper Atlas

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

Studying Very Low Resolution Recognition Using Deep Networks

Zhangyang Wang, Shiyu Chang, Yingzhen Yang, Ding Liu, Thomas S. Huang

CVPR 2016 · Label-efficient and robust learning › Recognition and long-tail · approach: Margin-based face recognition losses · 228 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 SR pre-training using HR training imagesBasic CNN · CIFAR-10top-1 classification error (%): 27.47 -> 25.15; 2.32%Improved
Add SR pre-training using HR training imagesBasic CNN · CIFAR-100top-1 classification error (%): 3.45%; 46.50%Improved
Add LR-HR feature transferModel II · CIFAR-10 and CIFAR-100top-1 classification error (%): 21.72%; 43.03%; 3.43%; 3.47%Improved
Replace full LR-HR feature coupling with partial couplingModel III · CIFAR-10 and CIFAR-100top-1 classification error (%): 18.77; 41.64; 2.95; 4.86Improved
Increase the third-layer coupled ratio from 0.75 to 1.00 while holding the first two at 0.50 (ablation)Partially coupled network · CIFAR-10top-1 classification error (%): 19.91% -> 20.72%Hurt
Use Huber loss instead of MSE in SR pre-training (ablation)Model IV · CIFAR-10 and CIFAR-100, cleantop-1 classification error (%): 18.77 -> 18.81; 41.64 -> 41.70No clear effect
Use Huber loss instead of MSE in SR pre-training (ablation)Model IV · CIFAR-10 and CIFAR-100 with 5%, 10%, and 15% corruptiontop-1 classification error (%): 19.99 -> 19.04; 23.03 -> 20.28; 25.05 -> 22.24; 43.43 -> 42.18; 46.20 -> 43.79; 49.49 -> 46.18Improved
Add SR pre-training, LR-HR transfer, and partial couplingVLRR face identification model · UCCS face identificationtop-1 accuracy: 40.97%@top-1; around 10%@top-1 over the simplest baselineImproved
Add Huber loss in pre-training (ablation)Model IV · UCCS face identificationtop-1 accuracy: 1.06%@top-1Improved
Add Huber loss in pre-training (ablation)Model IV · SVHNtop-1 accuracy: 4.05%@top-1Improved
Add LR-HR transfer (ablation)Model II · SVHNtop-1 accuracy: 4.03%@top-1Improved
Use Huber loss in pre-training (ablation)Model IV · VFR font recognitionrecognition errorNo clear effect

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