Robust 3D Face Alignment with Multi-Path Neural Architecture Search
This addresses the problem of inconsistent 3D face alignment across poses for computer vision applications, representing an incremental improvement over existing deep learning methods.
The paper tackles the challenge of 3D face alignment by using Neural Architecture Search (NAS) to automatically design networks, achieving superior performance on three benchmarks for both sparse and dense alignment.
3D face alignment is a very challenging and fundamental problem in computer vision. Existing deep learning-based methods manually design different networks to regress either parameters of a 3D face model or 3D positions of face vertices. However, designing such networks relies on expert knowledge, and these methods often struggle to produce consistent results across various face poses. To address this limitation, we employ Neural Architecture Search (NAS) to automatically discover the optimal architecture for 3D face alignment. We propose a novel Multi-path One-shot Neural Architecture Search (MONAS) framework that leverages multi-scale features and contextual information to enhance face alignment across various poses. The MONAS comprises two key algorithms: Multi-path Networks Unbiased Sampling Based Training and Simulated Annealing based Multi-path One-shot Search. Experimental results on three popular benchmarks demonstrate the superior performance of the MONAS for both sparse alignment and dense alignment.