Multimodal Deep Autoencoder for Human Pose Recovery

TLDR

Video-based human pose recovery typically retrieves poses from image features assuming a linear mapping between 2D images and 3D poses, yet this relationship is inherently nonlinear, limiting recovery performance. This study proposes a novel pose recovery method that employs a nonlinear mapping via a multi‑layered deep neural network. The method fuses multimodal features by constructing a low‑rank hypergraph Laplacian, obtaining a unified feature representation through eigen‑decomposition, and then trains a multi‑layer deep neural network via back‑propagation to learn a nonlinear mapping from 2D images to 3D poses. Experiments on three datasets show a 20–25 % reduction in recovery error, demonstrating the method’s effectiveness.

Abstract

Video-based human pose recovery is usually conducted by retrieving relevant poses using image features. In the retrieving process, the mapping between 2D images and 3D poses is assumed to be linear in most of the traditional methods. However, their relationships are inherently non-linear, which limits recovery performance of these methods. In this paper, we propose a novel pose recovery method using non-linear mapping with multi-layered deep neural network. It is based on feature extraction with multimodal fusion and back-propagation deep learning. In multimodal fusion, we construct hypergraph Laplacian with low-rank representation. In this way, we obtain a unified feature description by standard eigen-decomposition of the hypergraph Laplacian matrix. In back-propagation deep learning, we learn a non-linear mapping from 2D images to 3D poses with parameter fine-tuning. The experimental results on three data sets show that the recovery error has been reduced by 20%-25%, which demonstrates the effectiveness of the proposed method.