School of Computer Science and Technology, Zhejiang Sci-Tech University, Hangzhou 310018, China.
Sensors (Basel). 2023 Aug 22;23(17):7299. doi: 10.3390/s23177299.
In the field of human pose estimation, heatmap-based methods have emerged as the dominant approach, and numerous studies have achieved remarkable performance based on this technique. However, the inherent drawbacks of heatmaps lead to serious performance degradation in methods based on heatmaps for smaller-scale persons. While some researchers have attempted to tackle this issue by improving the performance of small-scale persons, their efforts have been hampered by the continued reliance on heatmap-based methods. To address this issue, this paper proposes the SSA Net, which aims to enhance the detection accuracy of small-scale persons as much as possible while maintaining a balanced perception of persons at other scales. SSA Net utilizes HRNetW48 as a feature extractor and leverages the TDAA module to enhance small-scale perception. Furthermore, it abandons heatmap-based methods and instead adopts coordinate vector regression to represent keypoints. Notably, SSA Net achieved an of 77.4% on the COCO Validation dataset, which is superior to other heatmap-based methods. Additionally, it achieved highly competitive results on the Tiny Validation and MPII datasets as well.
在人体姿态估计领域，基于热图的方法已经成为主流，许多研究基于该技术取得了显著的性能。然而，热图固有的缺陷导致基于热图的方法在小尺度人员方面的性能严重下降。虽然一些研究人员试图通过提高小尺度人员的性能来解决这个问题，但他们的努力受到了继续依赖基于热图的方法的阻碍。为了解决这个问题，本文提出了 SSA Net，旨在在保持对其他尺度人员的平衡感知的同时，尽可能提高小尺度人员的检测精度。SSA Net 使用 HRNetW48 作为特征提取器，并利用 TDAA 模块来增强小尺度感知。此外，它放弃了基于热图的方法，转而采用坐标向量回归来表示关键点。值得注意的是，SSA Net 在 COCO Validation 数据集上的准确率达到了 77.4%，优于其他基于热图的方法。此外，它在 Tiny Validation 和 MPII 数据集上也取得了极具竞争力的结果。
