Multi-person pose estimation (MPPE), which aims to locate keypoints for all
persons in the frames, is an active research branch of computer vision.
Variable human poses and complex scenes make MPPE dependent on both local
details and global structures, and the absence of them may cause keypoint
feature misalignment. In this case, high-order spatial interactions that can
effectively link the local and global information of features are particularly
important. However, most methods do not have spatial interactions, and a few
methods have low-order spatial interactions but they are difficult to achieve a
good balance between accuracy and complexity. To address the above problems, a
Dual-Residual Spatial Interaction Network (DRSI-Net) for MPPE with high
accuracy and low complexity is proposed in this paper. DRSI-Net recursively
performs residual spatial information interactions on neighbor features, so
that more useful spatial information can be retained and more similarities can
be obtained between shallow and deep extracted features. The channel and
spatial dual attention mechanism introduced in the multi-scale feature fusion
also helps the network to adaptively focus on features relevant to target
keypoints and further refine generated poses. At the same time, by optimizing
interactive channel dimensions and dividing gradient flow, the spatial
interaction module is designed to be lightweight, which reduces the complexity
of the network. According to the experimental results on the COCO dataset, the
proposed DRSI-Net outperforms other state-of-the-art methods in both accuracy
and complexity.

为了解决多人姿势估计中局部和全局信息不匹配的问题，本文提出了一种高精度低复杂度的 Dual-Residual Spatial Interaction Network (DRSI-Net)。通过在邻居特征上递归地进行残差空间信息交互，保留更有用的空间信息并在浅层和深层特征之间得到更多相似性。多尺度特征融合中引入的通道和空间双重关注机制有助于网络自适应地关注与目标关键点相关的特征并进一步改进生成的姿势。同时，通过优化交互通道维度和梯度流的分割，设计了轻量级的空间交互模块，从而减少了网络的复杂性。根据在 COCO 数据集上的实验结果，DRSI-Net 在精确性和复杂性方面均优于其他最先进的方法。