AI Paper Atlas

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

Improving Face Recognition from Hard Samples via Distribution Distillation Loss

Yuge Huang, Pengcheng Shen, Ying Tai, Shaoxin Li, Xiaoming Liu, Jilin Li, Feiyue Huang, Rongrong Ji

ECCV 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Fair and quality-aware face recognition · 69 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
replace conventional finetuning with DDLArcface (CASIA+R50) · SCface averagerank-1 accuracy (%): 86.3 -> 94.4Improved
add order loss to KL loss (ablation)DDL · SCface averagerank-1 accuracy (%): 89.6 -> 94.4Improved
add KL loss to order loss (ablation)DDL · SCface averagerank-1 accuracy (%): 92.4 -> 94.4Improved
use KL divergence instead of EMD (ablation)DDL · SCface averagerank-1 accuracy (%): 90.5 -> 94.4Improved
use KL divergence instead of JS divergence (ablation)DDL · SCface averagerank-1 accuracy (%): 93.4 -> 94.4Improved
use hard negative mining instead of random negative selection (ablation)DDL · SCface averagerank-1 accuracy (%): 90.6 -> 94.4Improved
use specific student distributions per resolution instead of one mixture (ablation)DDL · SCface averagerank-1 accuracy (%): 92.1 -> 94.4Improved
mixture-trained DDL versus conventional finetuningArcface-FT · SCface averagerank-1 accuracy (%): 86.3 -> 92.1Improved
add DDL on the hardest low-resolution setting d1Arcface-FT (CASIA+R50) · SCface d1rank-1 accuracy (%): 67.3 -> 86.8Improved
add DDL on easy d3 imagesArcface-FT (CASIA+R50) · SCface d3rank-1 accuracy (%): 98.0 -> 98.3No clear effect
finetune with DDL on pose hard samplesArcface (VGG+R100) · CPLFWverification accuracy (%): 93.13 -> 93.43Improved
finetune with DDL on pose hard samplesArcface (VGG+R100) · LFWverification accuracy (%): 99.62 -> 99.68No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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