AI Paper Atlas

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

DeCaFA: Deep Convolutional Cascade for Face Alignment in the Wild

Arnaud Dapogny, Kevin Bailly, Matthieu Cord

ICCV 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Margin-based face recognition losses · 97 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
reinject whole input image (F3 instead of F2 fusion) (ablation)DeCaFA · 300W-Challengingmean error (%): 5.45 -> 5.38Improved
reinject whole input image (F3 instead of F2 fusion) (ablation)DeCaFA · WFLW-Posemean error (%): 8.78 -> 8.39Improved
use F5 fusion instead of basic deep F1 fusion (ablation)DeCaFA · Average over 5 subsetsmean error (%): 4.83 -> 4.69Improved
use F4 fusion instead of F1 fusion (ablation)DeCaFA · Average over 5 subsetsmean error (%): 4.83 -> 4.80No clear effect
fuse with attention maps (F2 instead of F1) (ablation)DeCaFA · Average over 5 subsetsmean error (%): 4.83 -> 5.04Hurt
chain transfer layers instead of independent ones (ablation)DeCaFA · Average over 5 subsetsmean error (%): 4.75 -> 4.69Improved
increasing intermediate supervision weights instead of constant (ablation)DeCaFA · Average over 5 subsetsmean error (%): 4.91 -> 4.69Improved
increasing intermediate supervision weights instead of decreasing (ablation)DeCaFA · Average over 5 subsetsmean error (%): 5.05 -> 4.69Improved
increasing intermediate supervision weights instead of decreasing (ablation)DeCaFA · WFLW-Posemean error (%): 9.06 -> 8.11Improved
add CelebA and WFLW auxiliary data to 300W training (ablation)DeCaFA · 300W (full)mean error (%): 3.69 -> 3.39Improved
add 300W and CelebA auxiliary data to WFLW training (ablation)DeCaFA · WFLW allmean error (%): 5.01 -> 4.62Improved
add coarsely annotated CelebA (5 landmarks) data when training on few fine-annotated images (ablation)DeCaFA · 300W and WFLW with 3%/15% and 1%/6% of train setmean error (%)Improved

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