Enhancing Bronchoscopy Depth Estimation through Synthetic-to-Real Domain Adaptation
This work addresses a domain-specific challenge for medical imaging, providing an incremental improvement in depth estimation for bronchoscopy.
The paper tackles the problem of monocular depth estimation in bronchoscopic images, which lacks labeled data, by proposing a synthetic-to-real domain adaptation framework that improves depth prediction on real footage compared to training only on synthetic data.
Monocular depth estimation has shown promise in general imaging tasks, aiding in localization and 3D reconstruction. While effective in various domains, its application to bronchoscopic images is hindered by the lack of labeled data, challenging the use of supervised learning methods. In this work, we propose a transfer learning framework that leverages synthetic data with depth labels for training and adapts domain knowledge for accurate depth estimation in real bronchoscope data. Our network demonstrates improved depth prediction on real footage using domain adaptation compared to training solely on synthetic data, validating our approach.