arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.03100cs.CV
收藏

HIH:通过Heatmap中的Heatmap实现更精确的面部对齐

HIH: Towards More Accurate Face Alignment via Heatmap in Heatmap

Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);University of Chinese Academy of Sciences(中国科学院大学);Baidu Inc.(百度公司)

Xing Lan, Qinghao Hu, Qiang Chen, Jian Xue, Jian Cheng

摘要:

基于热图的回归克服了直接坐标回归缺乏空间和上下文信息的问题,并彻底改变了人脸对齐的任务。然而,它遭受的量化误差所造成的忽略亚像素坐标在图像去噪和网络降采样。在本文中,我们首先定量分析了量化误差的基准,占1/3以上的整体预测误差的最先进的方法。为了解决这个问题,我们提出了一种新的Heatmap In Heatmap(HIH)表示和一种坐标软分类(CSC)方法,它们无缝集成到经典的沙漏网络中。HIH表示使用嵌套的热图来共同表示坐标标签:一个称为整数热图的热图表示整数坐标,另一个称为十进制热图的热图表示子像素坐标。十进制热图的范围构成相应整数热图中的一个像素。此外,我们将偏移回归问题转化为一个区间分类问题,CSC将像素点的置信度视为区间的概率。同时,应用分布损失的CSC利用从高斯分布函数生成的软标签来指导偏移热图训练,这使得更容易学习坐标偏移的分布。在具有挑战性的基准数据集上进行的大量实验表明,我们的HIH可以实现最先进的结果。特别是,我们的HIH在WFLW上达到4.08 NME(归一化平均误差),在COFW上达到3.21,这大大超过了以前的方法。

英文摘要:

Heatmap-based regression overcomes the lack of spatial and contextual information of direct coordinate regression, and has revolutionized the task of face alignment. Yet it suffers from quantization errors caused by neglecting subpixel coordinates in image resizing and network downsampling. In this paper, we first quantitatively analyze the quantization error on benchmarks, which accounts for more than 1/3 of the whole prediction errors for state-of-the-art methods. To tackle this problem, we propose a novel Heatmap In Heatmap(HIH) representation and a coordinate soft-classification (CSC) method, which are seamlessly integrated into the classic hourglass network. The HIH representation utilizes nested heatmaps to jointly represent the coordinate label: one heatmap called integer heatmap stands for the integer coordinate, and the other heatmap named decimal heatmap represents the subpixel coordinate. The range of a decimal heatmap makes up one pixel in the corresponding integer heatmap. Besides, we transfer the offset regression problem to an interval classification task, and CSC regards the confidence of the pixel as the probability of the interval. Meanwhile, CSC applying the distribution loss leverage the soft labels generated from the Gaussian distribution function to guide the offset heatmap training, which makes it easier to learn the distribution of coordinate offsets. Extensive experiments on challenging benchmark datasets demonstrate that our HIH can achieve state-of-the-art results. In particular, our HIH reaches 4.08 NME (Normalized Mean Error) on WFLW, and 3.21 on COFW, which exceeds previous methods by a significant margin.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑