AI Paper Atlas

Research map / Video / Video segmentation and tracking

Learning Discriminative Aggregation Network for Video-Based Face Recognition

Yongming Rao, Ji Lin, Jiwen Lu, Jie Zhou

ICCV 2017 · Video › Video segmentation and tracking · approach: Deepfake video detection · 56 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
add DAN aggregation before feature extractionCNN dense features · YTFverification accuracy (%): 93.16 -> 94.28Improved
add DAN aggregation before feature extractionCNN dense features · PaSC controlverification rate at FAR 0.01 (%): 90.78 -> 92.06Improved
add DAN aggregation before feature extractionCNN dense features · PaSC handheldverification rate at FAR 0.01 (%): 78.67 -> 80.33Improved
add DAN aggregation before feature extractionCNN dense features · YTCclassification accuracy (%): 96.79 -> 97.32Improved
use DAN aggregation instead of random sampling of 10 frames (ablation)Random + CNN · YTFverification accuracy (%): 92.80 -> 94.28Improved
use DAN aggregation instead of hierarchical pooling into 10 frames (ablation)Hierarchical Pooling · YTFverification accuracy (%): 93.15 -> 94.28Improved
use DAN aggregation instead of random sampling (ablation)Random + CNN · PaSC controlverification rate at FAR 0.01 (%): 89.12 -> 92.06Improved
use DAN aggregation instead of hierarchical pooling (ablation)Hierarchical Pooling · PaSC handheldverification rate at FAR 0.01 (%): 78.23 -> 80.33Improved
use DAN aggregation instead of random sampling (ablation)Random + CNN · YTCclassification accuracy (%): 96.63 -> 97.32Improved
aggregate 200 frames to 10 images with DAN instead of dense extractionCNN dense features · YTFruntime (ms): 819.7 -> 126.1Improved

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