Segment-based Methods for Facial Attribute Detection from Partial Faces

TL;DR

This paper introduces SPLITFACE, a deep learning method that improves facial attribute detection in partially occluded faces by leveraging facial segments and a committee decision approach, outperforming existing methods on occluded face datasets.

Contribution

The paper presents SPLITFACE, a novel deep convolutional neural network architecture that effectively detects facial attributes from partial faces by using segment-based predictions and committee machine techniques.

Findings

SPLITFACE outperforms recent methods on occluded face datasets.

Segment-based predictions improve attribute detection accuracy.

Committee decision strategies enhance overall performance.

Abstract

State-of-the-art methods of attribute detection from faces almost always assume the presence of a full, unoccluded face. Hence, their performance degrades for partially visible and occluded faces. In this paper, we introduce SPLITFACE, a deep convolutional neural network-based method that is explicitly designed to perform attribute detection in partially occluded faces. Taking several facial segments and the full face as input, the proposed method takes a data driven approach to determine which attributes are localized in which facial segments. The unique architecture of the network allows each attribute to be predicted by multiple segments, which permits the implementation of committee machine techniques for combining local and global decisions to boost performance. With access to segment-based predictions, SPLITFACE can predict well those attributes which are localized in the visible parts of the face, without having to rely on the presence of the whole face. We use the CelebA and LFWA facial attribute datasets for standard evaluations. We also modify both datasets, to occlude the faces, so that we can evaluate the performance of attribute detection algorithms on partial faces. Our evaluation shows that SPLITFACE significantly outperforms other recent methods especially for partial faces.