Estimation of Ego-motion Velocities from Single Static Images

Abstract

Velocity estimation based on visual information is a well-researched topic. Traditional approaches usually rely on how a given feature or features change between successive images in a sequence. However, a single static image might contain motion information that could potentially be leveraged to estimate the optical flow. It can be hypothesized that motion blur and context of the scene are two sources of motion information in static images. This research work has two main goals, one is to investigate the prospect of using a learning-based framework to model a mapping directly to camera ego-motion velocity. The second goal is to analyze the contributing features in learning such a mapping. Experiments show that the model is able to learn velocity based on context of the scene but performs better when input images contain motion blur.