DeepAI AI Chat
Log In Sign Up

Human Pose Forecasting via Deep Markov Models

by   Sam Toyer, et al.
Australian National University

Human pose forecasting is an important problem in computer vision with applications to human-robot interaction, visual surveillance, and autonomous driving. Usually, forecasting algorithms use 3D skeleton sequences and are trained to forecast for a few milliseconds into the future. Long-range forecasting is challenging due to the difficulty of estimating how long a person continues an activity. To this end, our contributions are threefold: (i) we propose a generative framework for poses using variational autoencoders based on Deep Markov Models (DMMs); (ii) we evaluate our pose forecasts using a pose-based action classifier, which we argue better reflects the subjective quality of pose forecasts than distance in coordinate space; (iii) last, for evaluation of the new model, we introduce a 480,000-frame video dataset called Ikea Furniture Assembly (Ikea FA), which depicts humans repeatedly assembling and disassembling furniture. We demonstrate promising results for our approach on both Ikea FA and the existing NTU RGB+D dataset.


page 1

page 4

page 7


Deep Convolutional Poses for Human Interaction Recognition in Monocular Videos

Human interaction recognition is a challenging problem in computer visio...

Forecasting Human Dynamics from Static Images

This paper presents the first study on forecasting human dynamics from s...

A generic diffusion-based approach for 3D human pose prediction in the wild

3D human pose forecasting, i.e., predicting a sequence of future human 3...

Action-Agnostic Human Pose Forecasting

Predicting and forecasting human dynamics is a very interesting but chal...

Toward Reliable Human Pose Forecasting with Uncertainty

Recently, there has been an arms race of pose forecasting methods aimed ...

Towards Better Long-range Time Series Forecasting using Generative Forecasting

Long-range time series forecasting is usually based on one of two existi...

Pose Forecasting in Industrial Human-Robot Collaboration

Pushing back the frontiers of collaborative robots in industrial environ...