不用Soft-argmax也能实现顶尖人脸关键点检测效果
Heatmap Regression without Soft-Argmax for Facial Landmark Detection
- 改用结构化预测框架设计新训练目标，绕过不可导的Soft-argmax
- 在三个数据集上达到当前最好性能，训练速度提升2.2倍
- 适合追求高效训练和简洁模型架构的研究者
人脸关键点检测是计算机视觉中的重要任务，广泛应用于姿态估计、表情分析、人脸换脸等。基于热图回归的方法已广泛用于实现先进性能，其通常通过在热图上取argmax来预测关键点位置。由于argmax不可导，现有方法采用可微近似Soft-argmax以支持端到端训练。本文重新审视这一长期做法，证明Soft-argmax并非唯一路径。我们提出一种基于经典结构化预测框架的替代训练目标。实验表明，该方法在三个基准数据集（WFLW、COFW、300W）上均达到领先性能，训练收敛速度比基线快2.2倍，同时保持或优于现有精度。代码已开源：https://github.com/ca-joe-yang/regression-without-softarg。
原文摘要 · Abstract (English)
Facial landmark detection is an important task in computer vision with numerous applications, such as head pose estimation, expression analysis, face swapping, etc. Heatmap regression-based methods have been widely used to achieve state-of-the-art results in this task. These methods involve computing the argmax over the heatmaps to predict a landmark. Since argmax is not differentiable, these methods use a differentiable approximation, Soft-argmax, to enable end-to-end training on deep-nets. In this work, we revisit this long-standing choice of using Soft-argmax and demonstrate that it is not the only way to achieve strong performance. Instead, we propose an alternative training objective based on the classic structured prediction framework. Empirically, our method achieves state-of-the-art performance on three facial landmark benchmarks (WFLW, COFW, and 300W), converging 2.2x faster during training while maintaining better/competitive accuracy. Our code is available here: https://github.com/ca-joe-yang/regression-without-softarg.
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。