mlsanityData to 7 Oct 2026

Robust 3D Face Alignment with Multi-Path Neural Architecture Search

arXiv 2406.07873 ↗First posted 12 Jun 2024cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 20 of its 30 references that are in the map. First posted on arXiv 12 Jun 2024.

Builds on

20 of 30 references.

References · 20

  1. 2018DARTS: Differentiable Architecture Searchcited by 1,782
  2. 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
  3. 2016Stacked Hourglass Networks for Human Pose Estimationcited by 913
  4. 2019Once-for-All: Train One Network and Specialize it for Efficient Deploymentcited by 778
  5. 2019Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Setcited by 481
  6. 2019Single Path One-Shot Neural Architecture Search with Uniform Samplingcited by 411
  7. 2020Towards Fast, Accurate and Stable 3D Dense Face Alignmentcited by 219
  8. 2017SMASH: One-Shot Model Architecture Search through HyperNetworkscited by 200
  9. 2018Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Networkcited by 195
  10. 2020BigNAS: Scaling Up Neural Architecture Search with Big Single-Stage Modelscited by 187
  11. 2018Nonlinear 3D Face Morphable Modelcited by 157
  12. 2019Towards High-fidelity Nonlinear 3D Face Morphable Modelcited by 91
8 more
  1. 2017Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN Regressioncited by 66
  2. 20173D Face Morphable Models "In-the-Wild"cited by 44
  3. 2023A Hierarchical Representation Network for Accurate and Detailed Face Reconstruction from In-The-Wild Imagescited by 43
  4. 2021SADRNet: Self-Aligned Dual Face Regression Networks for Robust 3D Dense Face Alignment and Reconstructioncited by 27
  5. 2017Dense Face Alignmentcited by 24
  6. 2017Faster Than Real-time Facial Alignment: A 3D Spatial Transformer Network Approach in Unconstrained Posescited by 19
  7. 2018Hierarchical binary CNNs for landmark localization with limited resourcescited by 7
  8. 2023DSFNet: Dual Space Fusion Network for Occlusion-Robust 3D Dense Face Alignmentcited by 6

Built on by

Not cited yet.