search

akshitagupta15june / Face-X

Demonstration of different algorithms and operations on faces. Star the repo⭐
https://discord.gg/QapWBRZbVe
MIT License
697 stars 598 forks source link

Feature request : Multi-Modal Face Recognition System #1822

Open UTSAVS26 opened 3 weeks ago

UTSAVS26 commented 3 weeks ago

Is your feature request related to a problem? Please describe.
The problem revolves around the need for a reliable face recognition system capable of overcoming challenges such as variations in lighting conditions, facial expressions, and angles, as well as similarities between different individuals' facial features. Existing face recognition approaches may not adequately address these challenges.

Describe the solution you'd like
The desired solution is to develop a face recognition system utilizing various pre-trained models such as Xception, ResNet50, and VGG16. These models will undergo modifications using data augmentation techniques to enhance their accuracy under diverse conditions. Evaluation of each model's performance will be conducted using metrics like accuracy scores, loss, and confusion matrices to determine the most suitable model for the task.

Describe alternatives you've considered
Considered alternatives include exploring different network architectures, employing alternative data augmentation techniques, and modifying existing models to improve the face recognition system's accuracy. Additionally, considering alternative datasets and exploring modifications to existing models are also viable options.

Approach to be followed (optional)
The approach involves:

  1. Utilizing multiple pre-trained models for face recognition.
  2. Applying data augmentation techniques to enhance model accuracy.
  3. Comparing model performance using various evaluation metrics.
  4. Conducting Exploratory Data Analysis (EDA) to understand dataset characteristics.
  5. Documenting the work performed comprehensively using a README file.

Additional context
No additional context provided.

Full Name: Utsav Singhal GitHub Profile Link: https://github.com/UTSAVS26 Participant ID (If not, then put NA): Contributor What is your participant role? (Mention the Open Source Program name. Eg. HRSoC, GSSoC, GSOC etc.): SSOC

github-actions[bot] commented 3 weeks ago

Hello @UTSAVS26, Thank you for generating an issue to this project! Please wait while we get back to you.

akshitagupta15june commented 3 weeks ago

Go for it! and add your implementation inside face recognition folder