Low-latency Visual Servoing with Event-cameras

The dataset can be used to test your event-based tracking algorithm in open-loop.

In the experiment, a static event-camera (640x480 pixels) placed in front of an Alienware laptop displaying videos that recorded different motion types and target speeds, including a cluttered moving background.

The dataset includes event-driven data and ground truth of 8 shapes: arrow, flag, oval, pacman, puzzle, square, star, and triangle. 

To import .log files containing events, we suggest bimvee Python library.

Specifically, use the functions to import .log files:

data = importIitYarp(filePathOrName=input_path)

Ground-truth .csv files have 5 columns, each one corresponding to a different measure: 

timestamp | u | v | theta | scale

u and v refer to the target position in the image plane [pix]