Jonathan Bitton
Data Science Fellows June 2020 Cohort
Abstract
Detect facial attributes & accessories on images and label them accordingly.
Challenges (at least two)
- Dealing with large datasets and files
- Accuracy on small attributes
- Trade-off between frame per second and Accuracy/Loss (choice between DL or ML models)
Achievements (according to KPIs)
- Code library with classes such as loadmodel (for pretrained models), summarize (for metrics), preprocessing ( for data augmentation and preprocess), Train…
- Application using Flask with inference on images
Future project development
- Training on more attributes
- Generating images with unique attribute
- Improving small attribute detection
- Working on Neural networks to reduce the running time VS Improving performance with Machine Learning
- Improving face detection models for not frontal images