P-STMO: Pre-Trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimation

TL;DR

This paper proposes P-STMO, a two-stage pre-trained model for 3D human pose estimation that leverages self-supervised masked pose modeling and a many-to-one frame aggregation, achieving state-of-the-art accuracy with reduced computational cost.

Contribution

The paper introduces a novel pre-training strategy with masked pose modeling and a many-to-one aggregation for efficient 3D human pose estimation.

Findings

Outperforms state-of-the-art methods on benchmarks

Achieves 42.1mm MPJPE on Human3.6M dataset

Reduces computational overhead and model parameters

Abstract

This paper introduces a novel Pre-trained Spatial Temporal Many-to-One (P-STMO) model for 2D-to-3D human pose estimation task. To reduce the difficulty of capturing spatial and temporal information, we divide this task into two stages: pre-training (Stage I) and fine-tuning (Stage II). In Stage I, a self-supervised pre-training sub-task, termed masked pose modeling, is proposed. The human joints in the input sequence are randomly masked in both spatial and temporal domains. A general form of denoising auto-encoder is exploited to recover the original 2D poses and the encoder is capable of capturing spatial and temporal dependencies in this way. In Stage II, the pre-trained encoder is loaded to STMO model and fine-tuned. The encoder is followed by a many-to-one frame aggregator to predict the 3D pose in the current frame. Especially, an MLP block is utilized as the spatial feature extractor in STMO, which yields better performance than other methods. In addition, a temporal downsampling strategy is proposed to diminish data redundancy. Extensive experiments on two benchmarks show that our method outperforms state-of-the-art methods with fewer parameters and less computational overhead. For example, our P-STMO model achieves 42.1mm MPJPE on Human3.6M dataset when using 2D poses from CPN as inputs. Meanwhile, it brings a 1.5-7.1 times speedup to state-of-the-art methods. Code is available at https://github.com/paTRICK-swk/P-STMO.