Tracking using Autoencoders and Manifolds

Supervision: Christian Rupprecht, Federico Tombari

In this project we want to explore the possibilities of using autoencoders to perform object tracking in video sequences. The object's bounding box is given in the first frame and needs to be tracked throughout the sequence. We would like to use autoencoders to encode the appearance of the object and to predict its future (location and appearance).
While some initial work has been done to set up the framework and provide basic tracking. The goal is now to explore how to enforce the local temporal linearity during training of the autoencoder.

For further information please contact: Christian Rupprecht