Deja Vu: Motion Prediction in Static Images

TL;DR

This paper introduces a novel method to predict motion from static images by learning from videos, extending structured random forests, and demonstrates its usefulness in various applications like action recognition and saliency detection.

Contribution

It presents the first approach to predict motion in single images using appearance-based learning and extends structured random forests with regression for this task.

Findings

Motion prediction from static images is feasible.

Predicted motion improves action recognition accuracy.

Motion cues aid in detecting unexpected events.

Abstract

This paper proposes motion prediction in single still images by learning it from a set of videos. The building assumption is that similar motion is characterized by similar appearance. The proposed method learns local motion patterns given a specific appearance and adds the predicted motion in a number of applications. This work (i) introduces a novel method to predict motion from appearance in a single static image, (ii) to that end, extends of the Structured Random Forest with regression derived from first principles, and (iii) shows the value of adding motion predictions in different tasks such as: weak frame-proposals containing unexpected events, action recognition, motion saliency. Illustrative results indicate that motion prediction is not only feasible, but also provides valuable information for a number of applications.