# Exploiting temporal context for 3D human pose estimation in the wild

## TL;DR

This paper introduces a bundle-adjustment-based method that leverages temporal context in videos to improve 3D human pose and mesh estimation, demonstrating significant accuracy gains on various datasets.

## Contribution

It presents a novel algorithm that exploits temporal information across video sequences for more accurate 3D human pose estimation, and introduces a large-scale dataset for training.

## Key findings

- Improved 3D pose accuracy on Human 3.6M and in-the-wild datasets.
- Retraining on the new dataset enhances performance on 3DPW and HumanEVA.
- The method effectively resolves ambiguities present in single-frame approaches.

## Abstract

We present a bundle-adjustment-based algorithm for recovering accurate 3D human pose and meshes from monocular videos. Unlike previous algorithms which operate on single frames, we show that reconstructing a person over an entire sequence gives extra constraints that can resolve ambiguities. This is because videos often give multiple views of a person, yet the overall body shape does not change and 3D positions vary slowly. Our method improves not only on standard mocap-based datasets like Human 3.6M -- where we show quantitative improvements -- but also on challenging in-the-wild datasets such as Kinetics. Building upon our algorithm, we present a new dataset of more than 3 million frames of YouTube videos from Kinetics with automatically generated 3D poses and meshes. We show that retraining a single-frame 3D pose estimator on this data improves accuracy on both real-world and mocap data by evaluating on the 3DPW and HumanEVA datasets.