# Low-Cost Transfer Learning of Face Tasks

## TL;DR

This paper explores how understanding and leveraging filters in pretrained face recognition networks can enable low-cost transfer learning for various face-related tasks like age, pose, and emotion recognition, without extensive retraining.

## Contribution

It introduces a method to analyze face network filters to identify transferable features, facilitating efficient transfer learning across multiple face tasks without additional training.

## Key findings

- Filters encode task-specific information
- Transfer learning can be achieved without retraining
- Method improves efficiency of face task learning

## Abstract

Do we know what the different filters of a face network represent? Can we use this filter information to train other tasks without transfer learning? For instance, can age, head pose, emotion and other face related tasks be learned from face recognition network without transfer learning? Understanding the role of these filters allows us to transfer knowledge across tasks and take advantage of large data sets in related tasks. Given a pretrained network, we can infer which tasks the network generalizes for and the best way to transfer the information to a new task.