Joint Flow: Temporal Flow Fields for Multi Person Tracking

TL;DR

This paper introduces Temporal Flow Fields, a novel vector field representation for multi-person pose tracking that predicts joint movements between frames, enabling effective online tracking with state-of-the-art accuracy.

Contribution

The work proposes a new temporal model using Temporal Flow Fields that predict joint movements, improving multi-person pose tracking accuracy and efficiency.

Findings

Achieves state-of-the-art multi-person pose tracking results.

Uses a small CNN to learn effective Temporal Flow Fields.

Demonstrates robustness across different pose estimation methods.

Abstract

In this work we propose an online multi person pose tracking approach which works on two consecutive frames $I_{t-1}$ and $I_t$. The general formulation of our temporal network allows to rely on any multi person pose estimation approach as spatial network. From the spatial network we extract image features and pose features for both frames. These features serve as input for our temporal model that predicts Temporal Flow Fields (TFF). These TFF are vector fields which indicate the direction in which each body joint is going to move from frame $I_{t-1}$ to frame $I_t$. This novel representation allows to formulate a similarity measure of detected joints. These similarities are used as binary potentials in a bipartite graph optimization problem in order to perform tracking of multiple poses. We show that these TFF can be learned by a relative small CNN network whilst achieving state-of-the-art multi person pose tracking results.