SCAF: Skip-Connections in Auto-encoder for Face Alignment with Few Annotated Data - Archive ouverte HAL Access content directly
Conference Papers Year : 2022

SCAF: Skip-Connections in Auto-encoder for Face Alignment with Few Annotated Data

Abstract

Supervised face alignment methods need large amounts of training data to achieve good performance in terms of accuracy and generalization. However face alignment datasets rarely exceed a few thousand samples making these methods prone to overfitting on the specific training dataset. Semi-supervised methods like TS 3 or 3FabRec have emerged to alleviate this issue by using labeled and unlabeled data during the training. In this paper we propose Skip-Connections in Auto-encoder for Face alignment (SCAF), we build on 3FabRec by adding skip-connections between the encoder and the decoder. These skip-connections lead to better landmark predictions, especially on challenging examples. We also apply for the first time active learning to the face alignment task and introduce a new acquisition function, the Negative Neighborhood Magnitude, specially designed to assess the quality of heatmaps. These two proposals show their effectiveness on several face alignment datasets when training with limited data.
Fichier principal
Vignette du fichier
SCAF_submitted (2).pdf (1.07 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03687091 , version 1 (03-06-2022)

Identifiers

Cite

Martin Dornier, Philippe-Henri Gosselin, Christian Raymond, Yann Ricquebourg, Bertrand Coüasnon. SCAF: Skip-Connections in Auto-encoder for Face Alignment with Few Annotated Data. ICIAP 2022 - International Conference on Image Analysis and Processing, May 2022, Lecce, Italy. pp.425-437, ⟨10.1007/978-3-031-06427-2_36⟩. ⟨hal-03687091⟩
52 View
141 Download

Altmetric

Share

Gmail Facebook X LinkedIn More