Self-Supervised Generative Adversarial Network for Depth Estimation in Laparoscopic Images
This addresses the challenge of depth estimation in computer-assisted surgery for medical applications, but it is incremental as it builds on existing unsupervised methods with specific improvements.
The paper tackles the problem of dense depth estimation in laparoscopic images, where per-pixel ground truth data is scarce, by proposing SADepth, a self-supervised method based on Generative Adversarial Networks. The result shows that SADepth outperforms recent state-of-the-art unsupervised methods by a large margin and reduces the gap between supervised and unsupervised approaches.
Dense depth estimation and 3D reconstruction of a surgical scene are crucial steps in computer assisted surgery. Recent work has shown that depth estimation from a stereo images pair could be solved with convolutional neural networks. However, most recent depth estimation models were trained on datasets with per-pixel ground truth. Such data is especially rare for laparoscopic imaging, making it hard to apply supervised depth estimation to real surgical applications. To overcome this limitation, we propose SADepth, a new self-supervised depth estimation method based on Generative Adversarial Networks. It consists of an encoder-decoder generator and a discriminator to incorporate geometry constraints during training. Multi-scale outputs from the generator help to solve the local minima caused by the photometric reprojection loss, while the adversarial learning improves the framework generation quality. Extensive experiments on two public datasets show that SADepth outperforms recent state-of-the-art unsupervised methods by a large margin, and reduces the gap between supervised and unsupervised depth estimation in laparoscopic images.