# Transfer learning for vision-based tactile sensing

## TL;DR

This paper presents a vision-based tactile sensing approach using deep learning to reconstruct force distributions from images, with a calibration method enabling knowledge transfer across sensors, reducing training data needs.

## Contribution

It introduces a scalable soft optical tactile sensor and a transfer learning calibration procedure for efficient force reconstruction.

## Key findings

- High-resolution force distribution reconstruction from images
- Calibration enables knowledge transfer across sensors
- Reduced training data requirements

## Abstract

Due to the complexity of modeling the elastic properties of materials, the use of machine learning algorithms is continuously increasing for tactile sensing applications. Recent advances in deep neural networks applied to computer vision make vision-based tactile sensors very appealing for their high-resolution and low cost. A soft optical tactile sensor that is scalable to large surfaces with arbitrary shape is discussed in this paper. A supervised learning algorithm trains a model that is able to reconstruct the normal force distribution on the sensor's surface, purely from the images recorded by an internal camera. In order to reduce the training times and the need for large datasets, a calibration procedure is proposed to transfer the acquired knowledge across multiple sensors while maintaining satisfactory performance.