# A Deep Cascade Network for Unaligned Face Attribute Classification

arXiv: https://arxiv.org/abs/1709.03851 · HTML: https://mlsanity.com/paper/1709.03851/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Sep 2017.

## Built on by (most cited first, 8 of 8)

- [CelebV-HQ: A Large-Scale Video Facial Attributes Dataset](https://mlsanity.com/paper/2207.12393/) (2022)
- [Logical Consistency and Greater Descriptive Power for Facial Hair Attribute Learning](https://mlsanity.com/paper/2302.11102/) (2023)
- [Improving Facial Attribute Recognition by Group and Graph Learning](https://mlsanity.com/paper/2105.13825/) (2021)
- [LogicNet: A Logical Consistency Embedded Face Attribute Learning Network](https://mlsanity.com/paper/2311.11208/) (2023)
- [TransFA: Transformer-based Representation for Face Attribute Evaluation](https://mlsanity.com/paper/2207.05456/) (2022)
- [Facial Data Minimization: Shallow Model as Your Privacy Filter](https://mlsanity.com/paper/2310.15590/) (2023)
- [Mask-Guided Multi-Task Network for Face Attribute Recognition](https://mlsanity.com/paper/2601.01408/) (2026)
- [FAR-AMTN: Attention Multi-Task Network for Face Attribute Recognition](https://mlsanity.com/paper/2601.01537/) (2026)

## Abstract (arXiv)

Humans focus attention on different face regions when recognizing face attributes. Most existing face attribute classification methods use the whole image as input. Moreover, some of these methods rely on fiducial landmarks to provide defined face parts. In this paper, we propose a cascade network that simultaneously learns to localize face regions specific to attributes and performs attribute classification without alignment. First, a weakly-supervised face region localization network is designed to automatically detect regions (or parts) specific to attributes. Then multiple part-based networks and a whole-image-based network are separately constructed and combined together by the region switch layer and attribute relation layer for final attribute classification. A multi-net learning method and hint-based model compression is further proposed to get an effective localization model and a compact classification model, respectively. Our approach achieves significantly better performance than state-of-the-art methods on unaligned CelebA dataset, reducing the classification error by 30.9%.