AI Paper Atlas

Research map / Video / Video segmentation and tracking

TALL: Thumbnail Layout for Deepfake Video Detection

Yuting Xu, Jian Liang, Gengyun Jia, Ziming Yang, Yanhao Zhang, Ran He

ICCV 2023 · Video › Video segmentation and tracking · approach: Deepfake video detection · 134 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 TALL thumbnail layout to Swin-BSwin-B · FF++(HQ) -> Celeb-DFAUC: 83.13 -> 90.79Improved
add TALL to ViT-B, ResNet50, EfficientNet-B4image-level backbones · FF++(HQ) -> Celeb-DFAUC: 82.33 -> 86.58 (ViT-B); 76.38 -> 80.90 (ResNet50); 78.19 -> 83.37 (EffNetB4)Improved
use TALL with 2x2 resize layout rather than 4x4 layout (ablation)TALL-Swin · FF++(HQ) -> Celeb-DFAUC: 80.18 -> 87.60Improved
resize frames instead of random crop for sub-images (ablation)TALL-Swin · FF++(HQ) -> Celeb-DFAUC: 78.55 -> 87.60Improved
keep forward temporal order of frames instead of random (ablation)TALL-Swin · FF++(HQ) -> Celeb-DFAUC: 85.85 -> 87.60Improved
use all four sub-images rather than dropping the last two (ablation)TALL-Swin · FF++(HQ) -> Celeb-DFAUC: 84.22 -> 87.60Improved
use TALL fixed-position mask instead of Cutout (ablation)TALL-Swin · FF++(HQ) -> Celeb-DFAUC: 89.33 -> 90.79Improved
add TALL mask over Mixup+CutMix (ablation)TALL-Swin · FF++(HQ) -> Celeb-DFAUC: 89.75 -> 90.79Improved
enlarge Swin window to (14,14,14,7) from (7,7,7,7) (ablation)TALL-Swin · FF++(HQ) -> Celeb-DFAUC: 85.60 -> 87.60Improved
enlarge windows further to (28,28,28,7) (ablation)TALL-Swin · FF++(HQ) -> Celeb-DFAUC: 87.60 -> 86.65Hurt
use full-resolution sub-images (224x224, 2x2) (ablation)TALL-Swin · FF++(HQ) -> Celeb-DFAUC: 87.60 -> 88.15No clear effect
apply TALL on low-quality compressed FF++ (LQ)TALL-Swin · FF++ (LQ)Acc.: 92.82 vs ISTVT 96.15Hurt

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