https://github.com/nhix00/egogesture_recognition

Project for Computer Vision course at Sapienza University of Rome
https://github.com/nhix00/egogesture_recognition

cv egogesture gesture-recognition lightning pytorch

Last synced: about 2 hours ago
JSON representation

Project for Computer Vision course at Sapienza University of Rome

• Host: GitHub
• URL: https://github.com/nhix00/egogesture_recognition
• Owner: Nhix00
• License: mit
• Created: 2024-08-20T14:10:56.000Z (about 1 month ago)
• Default Branch: main
• Last Pushed: 2024-08-20T15:17:35.000Z (about 1 month ago)
• Last Synced: 2024-09-22T06:02:16.402Z (5 days ago)
• Topics: cv, egogesture, gesture-recognition, lightning, pytorch
• Language: Python
• Homepage:
• Size: 2.15 MB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Benchmark of Light Weight Models for EgoGesture Recognition

![Static Badge](https://img.shields.io/badge/PyTorch-white?logo=pytorch&link=https%3A%2F%2Fpytorch.org%2Fdocs%2Fstable%2Findex.html)

![Static Badge](https://img.shields.io/badge/PyTorch_Lightning-%23792EE5?logo=lightning&link=https%3A%2F%2Flightning.ai%2Fdocs%2Foverview%2Fgetting-started)

## Introduction

The proposed work primarily concentrates on offline gesture recognition as detailed in the [paper](https://doi.org/10.48550/arXiv.1901.10323). Our objective is to benchmark several lightweight models, such as ShuffleNet, and compare their performance to the heavier ResNeXt model featured in the study.

## Dataset

In this work we used **Egogesture** dataset that is a multi-modal large scale dataset for egocentric hand gesture recognition. This dataset provides the test-bed not only for gesture classification in segmented data but also for gesture detection in continuous data.

## Models

- **MobileNet**: Optimized for mobile and embedded devices with depth-wise separable convolutions for high-performance image recognition

- **ShuffleNet**: Minimizes computational complexity with channel shuffling, ideal for mobile and resource-constrained applications

We used pretrainet models on Jester dataset from this paper: [link](https://arxiv.org/pdf/1904.02422), they published their work on GitHub ([link](https://github.com/okankop/Efficient-3DCNNs))

## Results

The results of our work are shown in the next graphic.

![Benchmark of the tested models](./readme_assets/Benchmark.svg)

## Contacts

If you have any questions, suggestions, or feedback, we'd love to hear from you! Here's how you can reach out:

- Diego Barreto: [[email protected]](mailto:[email protected]?subject=[GitHub]%20EgoGesture_Recognition)

- Matteo Zacchino: [[email protected]](mailto:[email protected]?subject=[GitHub]%20EgoGesture_Recognition)