# Precise Facial Landmark Detection by Dynamic Semantic Aggregation Transformer

arXiv: https://arxiv.org/abs/2412.00740 · HTML: https://mlsanity.com/paper/2412.00740/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 24 references that are in the map.
- First posted on arXiv 1 Dec 2024.

## Builds on (most cited first, 8 of 8)

- [Dynamic Neural Networks: A Survey](https://mlsanity.com/paper/2102.04906/) (2021)
- [Fast Decoding in Sequence Models using Discrete Latent Variables](https://mlsanity.com/paper/1803.03382/) (2018)
- [Robust Face Alignment by Multi-order High-precision Hourglass Network](https://mlsanity.com/paper/2010.08722/) (2020)
- [Precise Facial Landmark Detection by Reference Heatmap Transformer](https://mlsanity.com/paper/2303.07840/) (2023)
- [AnchorFace: An Anchor-based Facial Landmark Detector Across Large Poses](https://mlsanity.com/paper/2007.03221/) (2020)
- [MobileFAN: Transferring Deep Hidden Representation for Face Alignment](https://mlsanity.com/paper/1908.03839/) (2019)
- [Heterogeneous Face Recognition via Face Synthesis with Identity-Attribute Disentanglement](https://mlsanity.com/paper/2206.04854/) (2022)
- [Reasoning Structural Relation for Occlusion-Robust Facial Landmark Localization](https://mlsanity.com/paper/2112.10087/) (2021)

## Built on by (most cited first, 6 of 6)

- [Proto-Former: Unified Facial Landmark Detection by Prototype Transformer](https://mlsanity.com/paper/2510.15338/) (2025)
- [FGTBT: Frequency-Guided Task-Balancing Transformer for Unified Facial Landmark Detection](https://mlsanity.com/paper/2601.12863/) (2026)
- [Supervision-by-Hallucination-and-Transfer: A Weakly-Supervised Approach for Robust and Precise Facial Landmark Detection](https://mlsanity.com/paper/2601.12919/) (2026)
- [FreqFLD: Towards All-in-One Facial Landmark Detection via Frequency Modulation](https://mlsanity.com/paper/2609.10278/) (2026)
- [Fine-Grained Visual Preprocessing and Dual-Stream Temporal Modeling for Multimodal Sentiment Analysis on Social Media](https://mlsanity.com/paper/2609.07010/) (2026)
- [FAHCD-Net: Frequency-Adaptive Heatmap-Conditional Diffusion Networks for Robust Facial Landmark Detection](https://mlsanity.com/paper/2609.16842/) (2026)

## Abstract (arXiv)

At present, deep neural network methods have played a dominant role in face alignment field. However, they generally use predefined network structures to predict landmarks, which tends to learn general features and leads to mediocre performance, e.g., they perform well on neutral samples but struggle with faces exhibiting large poses or occlusions. Moreover, they cannot effectively deal with semantic gaps and ambiguities among features at different scales, which may hinder them from learning efficient features. To address the above issues, in this paper, we propose a Dynamic Semantic-Aggregation Transformer (DSAT) for more discriminative and representative feature (i.e., specialized feature) learning. Specifically, a Dynamic Semantic-Aware (DSA) model is first proposed to partition samples into subsets and activate the specific pathways for them by estimating the semantic correlations of feature channels, making it possible to learn specialized features from each subset. Then, a novel Dynamic Semantic Specialization (DSS) model is designed to mine the homogeneous information from features at different scales for eliminating the semantic gap and ambiguities and enhancing the representation ability. Finally, by integrating the DSA model and DSS model into our proposed DSAT in both dynamic architecture and dynamic parameter manners, more specialized features can be learned for achieving more precise face alignment. It is interesting to show that harder samples can be handled by activating more feature channels. Extensive experiments on popular face alignment datasets demonstrate that our proposed DSAT outperforms state-of-the-art models in the literature.Our code is available at https://github.com/GERMINO-LiuHe/DSAT.