ac-freeman / dvs-fast-corners

Rust implementation and visualization of the Fast Corners algorithm for the Dynamic Vision Sensor (DVS) camera

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

dvs-fast-corners

This is a Rust implementation and visualization of the Fast Corners algorithm for the Dynamic Vision Sensor (DVS) camera.

Based on the work of:

E. Mueggler, C. Bartolozzi, D. Scaramuzza: Fast Event-based Corner Detection. British Machine Vision Conference (BMVC), London, 2017.

Find the original paper here.

Screenshot from 2023-04-17 17-01-36-1
Features detected over 1/60th of a second are marked with a +. Note the predominance around the corners of the object.

Setup and Usage

  • Install Rust
  • Clone this repository and cd into it
  • Run cargo run --release -- --input "/path/to/aedat4/file"
  • Log the detected features to a file with cargo run --features "feature-logging" --release -- --input "/path/to/aedat4/file"

Setup is simple, with no non-Rust requirements.

Out of the box, this works with .aedat4 files created with iniVation cameras. The default resolution is 346x260, but you can easily change this to suit your needs.

About

Rust implementation and visualization of the Fast Corners algorithm for the Dynamic Vision Sensor (DVS) camera

License:MIT License


Languages

Language:Rust 100.0%