GADS: A Super Lightweight Model for Head Pose Estimation

2025/04/22 by Menan Velayuthan, Asiri Gawesha, Velayuthan, Menan +9 · 2 citations
Computer Science · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Hand Gesture Recognition Systems #Human Pose and Action Recognition

paper · pdf · doi:10.48550/arxiv.2504.15751

openalex publication_date 2025/04/22 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

In human-computer interaction, head pose estimation profoundly influences application functionality. Although utilizing facial landmarks is valuable for this purpose, existing landmark-based methods prioritize precision over simplicity and model size, limiting their deployment on edge devices and in compute-poor environments. To bridge this gap, we propose Grouped Attention Deep Sets (GADS), a novel architecture based on the Deep Set framework. By grouping landmarks into regions and employing small Deep Set layers, we reduce computational complexity. Our multihead attention mechanism extracts and combines inter-group information, resulting in a model that is 7.5× smaller and executes 25× faster than the current lightest state-of-the-art model. Notably, our method achieves an impressive reduction, being 4321× smaller than the best-performing model. We introduce vanilla GADS and Hybrid-GADS (landmarks + RGB) and evaluate our models on three benchmark datasets -- AFLW2000, BIWI, and 300W-LP. We envision our architecture as a robust baseline for resource-constrained head pose estimation methods.

Citations

• img2pose: Face Alignment and Detection via 6DoF, Face Pose Estimation
• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• A Vector-based Representation to Enhance Head Pose Estimation
• WHENet: Real-time Fine-Grained Estimation for Wide Range Head Pose
• Deep Learning for 3D Point Clouds: A Survey
• Deep Learning for 3D Point Clouds: A Survey
• MediaPipe: A Framework for Building Perception Pipelines
• Improving Head Pose Estimation with a Combined Loss and Bounding Box Margin Adjustment
• Light-weight Head Pose Invariant Gaze Tracking
• Self-Attention with Relative Position Representations
• Every Smile is Unique: Landmark-Guided Diverse Smile Generation
• Face-from-Depth for Head Pose Estimation on Depth Images
• Fine-Grained Head Pose Estimation Without Keypoints
• Attention Is All You Need
• PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
• How far are we from solving the 2D & 3D Face Alignment problem? (and a dataset of 230,000 3D facial landmarks)
• Deep Sets
• KEPLER: Keypoint and Pose Estimation of Unconstrained Faces by Learning Efficient H-CNN Regressors
• PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
• POSEidon: Face-from-Depth for Driver Pose Estimation
• Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networks
• Recent Advances in Convolutional Neural Networks
• Recent advances in convolutional neural networks
• Face Alignment in Full Pose Range: A 3D Total Solution
• Attention-Based Models for Speech Recognition
• Neural Machine Translation by Jointly Learning to Align and Translate
• Gradient-based learning applied to document recognition

Related

• On the power of data augmentation for head pose estimation
• A lightweight 3D dense facial landmark estimation model from position map data
• From Depth Data to Head Pose Estimation: a Siamese approach
• Nose, eyes and ears: Head pose estimation by locating facial keypoints
• A Real-Time Multi-Task Learning System for Joint Detection of Face, Facial Landmark and Head Pose