轻量级头姿态估计模型，速度提升25倍，体积缩小4321倍。
GADS: A Super Lightweight Model for Head Pose Estimation
- 将面部关键点分组，用小规模深度集架构降低计算开销。
- 在三个基准数据集上达到领先精度，模型比最轻模型小7.5倍、快25倍。
- 适合边缘设备部署，尤其适合算力受限的实时交互场景。
在人机交互中，头姿态估计对功能实现至关重要。尽管基于面部关键点的方法具有价值，但现有方法更关注精度而非简洁性与模型尺寸，限制了其在边缘设备和计算资源匮乏环境中的部署。为此，我们提出新颖的分组注意力深度集（GADS）架构，基于深度集框架，将关键点分组并采用小型深度集层以降低计算复杂度。通过多头注意力机制提取并融合组间信息，使模型体积缩小至当前最轻模型的1/7.5，推理速度提升25倍；特别地，相比表现最佳模型，体积减少4321倍。我们提出了纯关键点的GADS与结合RGB图像的Hybrid-GADS，并在AFLW2000、BIWI和300W-LP三个基准数据集上进行了评估。本架构可作为资源受限环境下头姿态估计的稳健基线。
原文摘要 · Abstract (English)
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 \textbf{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 $7.5\times$ smaller and executes $25\times$ faster than the current lightest state-of-the-art model. Notably, our method achieves an impressive reduction, being $4321\times$ 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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。