# Tree-gated Deep Mixture-of-Experts For Pose-robust Face Alignment

arXiv: https://arxiv.org/abs/1910.09450 · HTML: https://mlsanity.com/paper/1910.09450/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Vision mixture-of-experts, one of 12 directions in Mixture-of-experts.
- Builds on 14 of its 30 references that are in the map.
- First posted on arXiv 21 Oct 2019.

## Placed in

- Mixture-of-experts / [Vision mixture-of-experts](https://mlsanity.com/moe/vision-mixture-of-experts/)

## Builds on (most cited first, 14 of 14)

- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Cross-stitch Networks for Multi-task Learning](https://mlsanity.com/paper/1604.03539/) (2016)
- [Learning Factored Representations in a Deep Mixture of Experts](https://mlsanity.com/paper/1312.4314/) (2013)
- [Fine-Grained Head Pose Estimation Without Keypoints](https://mlsanity.com/paper/1710.00925/) (2017)
- [Face Alignment in Full Pose Range: A 3D Total Solution](https://mlsanity.com/paper/1804.01005/) (2018)
- [Adaptive Neural Trees](https://mlsanity.com/paper/1807.06699/) (2018)
- [Improving Landmark Localization with Semi-Supervised Learning](https://mlsanity.com/paper/1709.01591/) (2017)
- [Learning Deep Representation for Face Alignment with Auxiliary Attributes](https://mlsanity.com/paper/1408.3967/) (2014)
- [Disentangling 3D Pose in A Dendritic CNN for Unconstrained 2D Face Alignment](https://mlsanity.com/paper/1802.06713/) (2018)
- [Decision Forests, Convolutional Networks and the Models in-Between](https://mlsanity.com/paper/1603.01250/) (2016)
- [Simultaneous Facial Landmark Detection, Pose and Deformation Estimation under Facial Occlusion](https://mlsanity.com/paper/1709.08130/) (2017)
- [Pose-Invariant Face Alignment with a Single CNN](https://mlsanity.com/paper/1707.06286/) (2017)
- [Face Alignment with Cascaded Semi-Parametric Deep Greedy Neural Forests](https://mlsanity.com/paper/1703.01597/) (2017)
- [Tree-gated Deep Regressor Ensemble For Face Alignment In The Wild](https://mlsanity.com/paper/1907.03248/) (2019)

## Built on by (most cited first, 5 of 5)

- [Objects are Different: Flexible Monocular 3D Object Detection](https://mlsanity.com/paper/2104.02323/) (2021)
- [Adversarial Mixture Of Experts with Category Hierarchy Soft Constraint](https://mlsanity.com/paper/2007.12349/) (2020)
- [Privileged Attribution Constrained Deep Networks for Facial Expression Recognition](https://mlsanity.com/paper/2203.12905/) (2022)
- [Using Deep Mixture-of-Experts to Detect Word Meaning Shift for TempoWiC](https://mlsanity.com/paper/2211.03466/) (2022)
- [FreqFLD: Towards All-in-One Facial Landmark Detection via Frequency Modulation](https://mlsanity.com/paper/2609.10278/) (2026)

## Abstract (arXiv)

Face alignment consists of aligning a shape model on a face image. It is an active domain in computer vision as it is a preprocessing for a number of face analysis and synthesis applications. Current state-of-the-art methods already perform well on "easy" datasets, with moderate head pose variations, but may not be robust for "in-the-wild" data with poses up to 90deg. In order to increase robustness to an ensemble of factors of variations (e.g. head pose or occlusions), a given layer (e.g. a regressor or an upstream CNN layer) can be replaced by a Mixture of Experts (MoE) layer that uses an ensemble of experts instead of a single one. The weights of this mixture can be learned as gating functions to jointly learn the experts and the corresponding weights. In this paper, we propose to use tree-structured gates which allows a hierarchical weighting of the experts (Tree-MoE). We investigate the use of Tree-MoE layers in different contexts in the frame of face alignment with cascaded regression, firstly for emphasizing relevant, more specialized feature extractors depending of a high-level semantic information such as head pose (Pose-Tree-MoE), and secondly as an overall more robust regression layer. We perform extensive experiments on several challenging face alignment datasets, demonstrating that our approach outperforms the state-of-the-art methods.