Lightweight Multi-Resolution Network for Human Pose Estimation

Human pose estimation aims to localize the body joints from image or video data. With the development of deep learning, pose estimation has become a hot research topic in the field of computer vision. In recent years, human pose estimation has achieved great success in multiple fields such as animation and sports. However, to obtain accurate positioning results, existing methods may suffer from large model sizes, a high number of parameters, and increased complexity, leading to high computing costs. In this paper, we propose a new lightweight feature encoder to construct a high -resolution network that reduces the number of parameters and lowers the computing cost. We also introduced a semantic enhancement module that improves global feature extraction and network performance by combining channel and spatial dimensions. Furthermore, we propose a dense connected spatial pyramid pooling module to compensate for the decrease in image resolution and information loss in the network. Finally, our method effectively reduces the number of parameters and complexity while ensuring high performance. Extensive experiments show that our method achieves a competitive performance while dramatically reducing the number of parameters, and operational complexity. Specifically, our method can obtain 89.9% AP score on MPII VAL, while the number of parameters and the complexity of operations were reduced by 41% and 36%, respectively.