ThermoCycleNet: Stereo-based Thermogram Labeling for Model Transition to Cycling

TL;DR

This paper presents ThermoCycleNet, a stereo-based thermogram labeling method that efficiently adapts deep neural networks from treadmill running to cycling in sports medicine using minimal manual annotations.

Contribution

It introduces a transfer learning approach that combines automatic labels with limited manual annotations to improve thermogram segmentation for different exercise modalities.

Findings

Fine-tuning with minimal manual data enhances network performance.

Automatic labeling accelerates adaptation to new use cases.

Method effectively transfers from treadmill to cycling thermogram analysis.

Abstract

Infrared thermography is emerging as a powerful tool in sports medicine, allowing assessment of thermal radiation during exercise and analysis of anatomical regions of interest, such as the well-exposed calves. Building on our previous advanced automatic annotation method, we aimed to transfer the stereo- and multimodal-based labeling approach from treadmill running to ergometer cycling. Therefore, the training of the semantic segmentation network with automatic labels and fine-tuning on high-quality manually annotated images has been examined and compared in different data set combinations. The results indicate that fine-tuning with a small fraction of manual data is sufficient to improve the overall performance of the deep neural network. Finally, combining automatically generated labels with small manually annotated data sets accelerates the adaptation of deep neural networks to new use cases, such as the transition from treadmill to bicycle.