Learning to Forecast Videos of Human Activity with Multi-granularity Models and Adaptive Rendering

TL;DR

This paper introduces a multi-granularity model combining hierarchical pose prediction and adaptive appearance rendering to forecast complex human activity videos, outperforming existing methods.

Contribution

It presents a novel hierarchical temporal model with dynamic interaction mechanisms and an adaptive appearance rendering network for improved video forecasting.

Findings

Generated videos surpass state-of-the-art methods in quality.

Model effectively handles complex multi-person activities.

Adaptive rendering improves appearance accuracy.

Abstract

We propose an approach for forecasting video of complex human activity involving multiple people. Direct pixel-level prediction is too simple to handle the appearance variability in complex activities. Hence, we develop novel intermediate representations. An architecture combining a hierarchical temporal model for predicting human poses and encoder-decoder convolutional neural networks for rendering target appearances is proposed. Our hierarchical model captures interactions among people by adopting a dynamic group-based interaction mechanism. Next, our appearance rendering network encodes the targets' appearances by learning adaptive appearance filters using a fully convolutional network. Finally, these filters are placed in encoder-decoder neural networks to complete the rendering. We demonstrate that our model can generate videos that are superior to state-of-the-art methods, and can handle complex human activity scenarios in video forecasting.