Fast Human Pose Estimation
This addresses the practical issue of poor scalability and cost-effectiveness for deploying pose estimation models, though it is incremental as it builds on existing distillation techniques.
The paper tackled the efficiency problem in human pose estimation by proposing a Fast Pose Distillation (FPD) model learning strategy, which trains a lightweight neural network to achieve rapid execution with low computational cost, outperforming state-of-the-art methods on MPII Human Pose and Leeds Sports Pose datasets.
Existing human pose estimation approaches often only consider how to improve the model generalisation performance, but putting aside the significant efficiency problem. This leads to the development of heavy models with poor scalability and cost-effectiveness in practical use. In this work, we investigate the under-studied but practically critical pose model efficiency problem. To this end, we present a new Fast Pose Distillation (FPD) model learning strategy. Specifically, the FPD trains a lightweight pose neural network architecture capable of executing rapidly with low computational cost. It is achieved by effectively transferring the pose structure knowledge of a strong teacher network. Extensive evaluations demonstrate the advantages of our FPD method over a broad range of state-of-the-art pose estimation approaches in terms of model cost-effectiveness on two standard benchmark datasets, MPII Human Pose and Leeds Sports Pose.