There is an increasing demand for lightweight multi-person pose estimation
for many emerging smart IoT applications. However, the existing algorithms tend
to have large model sizes and intense computational requirements, making them
ill-suited for real-time applications and deployment on resource-constrained
hardware. Lightweight and real-time approaches are exceedingly rare and come at
the cost of inferior accuracy. In this paper, we present EfficientHRNet, a
family of lightweight multi-person human pose estimators that are able to
perform in real-time on resource-constrained devices. By unifying recent
advances in model scaling with high-resolution feature representations,
EfficientHRNet creates highly accurate models while reducing computation enough
to achieve real-time performance. The largest model is able to come within 4.4%
accuracy of the current state-of-the-art, while having 1/3 the model size and
1/6 the computation, achieving 23 FPS on Nvidia Jetson Xavier. Compared to the
top real-time approach, EfficientHRNet increases accuracy by 22% while
achieving similar FPS with 1/3 the power. At every level, EfficientHRNet proves
to be more computationally efficient than other bottom-up 2D human pose
estimation approaches, while achieving highly competitive accuracy.

本研究提出了 EfficientHRNet，这是一系列轻量级多人姿势估计器，能够在资源受限设备上实时执行姿势估计操作。通过将模型缩放的最新进展与高分辨率特征表示的统一，EfficientHRNet 在降低计算量的同时创建高度准确的模型，最大的模型能够在 1/3 的计算量和 1/3 的功率的情况下，实现与现有最先进水平相近的精度。