The project utilized state-of-the-art technologies, leveraging deep learning frameworks such as
TensorFlow and PyTorch. The chosen frameworks offer pre-trained models suitable for face
tasks, streamlining the development process.
The team opted for a Convolutional Neural Network (CNN) based model due to its efficiency in
recognition tasks. Transfer learning was employed, utilizing a pre-trained model and fine-tuning
custom dataset to adapt it specifically for face detection.
Integration with Mobile Platforms
For seamless integration with Android and iOS, the team developed native modules using Android
(for Android) and Xcode (for iOS). This ensured optimal performance and native user experiences
The implementation focused on achieving real-time face detection, requiring optimization for
devices' limited resources. Model quantization and efficient memory management were key
address this challenge.