Fast 6-DoF Object Pose Estimation with Event Cameras

Objectives

1. Estimation of the 6-DoF Object Pose from input from an event-camera
2. Evaluation and comparison to state-of-the-art

Requirements

1. Selection of state-of-the-art network for 6-DoF pose estimation most compatible with the task and retraining. Modify the network as needed to make it suitable for the eventdriven data.
2. Preparation of dataset for training and testing the network. Refinement of training requirements after initial performance analysis
3. Evaluation of the 6-DoF pose results against state-of-the-art in terms of accuracy and run-time.