Event-based Gesture and Facial Expression Recognition: A Comparative Analysis

Codes for "Event-based Gesture and Facial Expression Recognition: A Comparative Analysis"
Explore our laboratory»

This repository contains instructions on evaluated/implemented event-based objects recognition methods for gesture and facial expression recognition task, as described in R. Verschae, I. Bugueno-Cordova "Event-based Gesture and Facial Expression Recognition: A Comparative Analysis". The paper can be found here

If you cite this article in an academic context, we appreciate a citation to the paper below:

@ARTICLE{,
  author={},
  journal={IEEE Access}, 
  title={}, 
  year={2023},
  volume={},
  number={},
  pages={},
  doi={}
}

Datasets

• IBM DVS Gesture 128: The dataset can be accessed through the official website, associated to the paper "A. Amir, et. al. "A Low Power, Fully Event-Based Gesture Recognition System," 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, 2017.". We thank the authors for their contribution.

• NavGesture: The dataset can be accessed through the official website, associated to the paper "Maro J-M, et. al., (2020) Event-Based Gesture Recognition With Dynamic Background Suppression Using Smartphone Computational Capabilities. Front. Neurosci. 14:275.". We thank the authors for their contribution.

• e-CK+: The novel and synthetic event-based facial expression recognition dataset emulated from the "The Extended Cohn-Kanade Dataset (CK+)" can be requested and accessed through this following website. We thank the original authors for their contribution.

• e-MMI: The novel and synthetic event-based facial expression recognition dataset emulated from the "MMI Facial Expression Database" can be requested and accessed through this following website. We thank the original authors for their contribution.

Codes, Requirements & Installation

• EST: Code, requirements and installation instructions provided in the authors repository, which corresponds to the implementation of the paper "D. Gehrig, et. al., "End-to-End Learning of Representations for Asynchronous Event-Based Data," 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Korea (South), 2019, pp. 5632-5642."

• Asynet: Code, requirements and installation instructions provided in the authors repository, which corresponds to the implementation of the paper "Messikommer, N., et. al., (2020). Event-Based Asynchronous Sparse Convolutional Networks. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, JM. (eds) Computer Vision – ECCV 2020. ECCV 2020. Lecture Notes in Computer Science(), vol 12353. Springer, Cham."

• ESTM: Code provided in this repository, specifically in the 'estm/' folder. Requirements and installation instructions are identical to EST.

Technical Details

For more information, please see the "Event-based Gesture and Facial Expression Recognition: A Comparative Analysis" website.