In the latest machine learning project, I have developed a powerful model for grouping similar faces, incorporating techniques from deep learning and computer vision. The custom dataset comprised 300 images, ensuring comprehensive representation of facial variations.The core architecture of the model was based on a sophisticated Siamese network, enabling robust feature extraction from both single and multiple faces within images. To enhance accuracy, pretrained models like FaceNet were integrated, and YOLOV8 and MTCNN detectors were used to extract precise facial embeddings. These embeddings were efficiently processed through the FaceNet model, yielding exceptional results for single faces. Interestingly, the FaceNet model faced challenges with multiple faces, while the Siamese network consistently outperformed, offering superior performance in both scenarios and delivering reliable clustering results.This remarkable project underscores the potential of combining state-of-the-art deep learning and computer vision techniques for practical applications such as facial recognition and clustering. The groundbreaking solutions showcased a passion for innovation and a keen dedication to advancing AI. #MachineLearning #DeepLearning #ComputerVision #FacialRecognition #SiameseNetworks #FaceNet #YOLOV8 #MTCNN #ClusteringAlgorithms #DataScience #AI #Innovation
-
Notifications
You must be signed in to change notification settings - Fork 0
ML model for grouping similar faces using cutting-edge deep learning and computer vision techniques. Custom dataset of 300 images captures comprehensive facial variations. Siamese network outperforms Face-Net, delivering reliable clustering results.
License
kushagr-8/Face-Clustering-Using-Siamese-Network-
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
ML model for grouping similar faces using cutting-edge deep learning and computer vision techniques. Custom dataset of 300 images captures comprehensive facial variations. Siamese network outperforms Face-Net, delivering reliable clustering results.
Topics
Resources
License
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published