MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Video
This work addresses the problem of insufficient spatio-temporal correlation learning in 3D human pose estimation for video analysis, offering a novel method that enhances accuracy for applications like motion capture and human-computer interaction.
The paper tackles 3D human pose estimation in video by proposing MixSTE, a seq2seq mixed spatio-temporal encoder that separately models temporal motion per joint and spatial correlation between joints, achieving state-of-the-art results with improvements of 10.9% P-MPJPE and 7.6% MPJPE on benchmarks.
Recent transformer-based solutions have been introduced to estimate 3D human pose from 2D keypoint sequence by considering body joints among all frames globally to learn spatio-temporal correlation. We observe that the motions of different joints differ significantly. However, the previous methods cannot efficiently model the solid inter-frame correspondence of each joint, leading to insufficient learning of spatial-temporal correlation. We propose MixSTE (Mixed Spatio-Temporal Encoder), which has a temporal transformer block to separately model the temporal motion of each joint and a spatial transformer block to learn inter-joint spatial correlation. These two blocks are utilized alternately to obtain better spatio-temporal feature encoding. In addition, the network output is extended from the central frame to entire frames of the input video, thereby improving the coherence between the input and output sequences. Extensive experiments are conducted on three benchmarks (Human3.6M, MPI-INF-3DHP, and HumanEva). The results show that our model outperforms the state-of-the-art approach by 10.9% P-MPJPE and 7.6% MPJPE. The code is available at https://github.com/JinluZhang1126/MixSTE.