Skeleton-based Gait Index Estimation with LSTMs

TL;DR

This paper introduces a novel LSTM-based system that estimates gait indices from skeleton sequences by reconstructing input features and using the reconstruction error as an indicator, outperforming recent gait analysis methods.

Contribution

It presents a new LSTM encoder-decoder framework that automatically extracts gait features and estimates gait indices from skeleton data.

Findings

Outperforms recent gait analysis methods on a large dataset

Uses reconstruction error as a weak gait index

Effective for long-term gait analysis

Abstract

In this paper, we propose a method that estimates a gait index for a sequence of skeletons. Our system is a stack of an encoder and a decoder that are formed by Long Short-Term Memories (LSTMs). In the encoding stage, the characteristics of an input are automatically determined and are compressed into a latent space. The decoding stage then attempts to reconstruct the input according to such intermediate representation. The reconstruction error is thus considered as a weak gait index. By combining such weak indices over a long-time movement, our system can provide a good estimation for the gait index. Our experiments on a large dataset (nearly one hundred thousand skeletons) showed that the index given by the proposed method outperformed some recent works on gait analysis.