3D endoscope system with AR display superimposing dense and wide-angle-of-view 3D points obtained by using micro pattern projector.

In recent years, augmented reality (AR) technologies have been widespread for supporting various kinds of tasks, by superimposing useful information on the users' view of the real environments. In endoscopic diagnosis, AR systems can be helpful as an aid in presenting information to endoscopists who have their hands full. In this paper, we propose a system that can superimpose shapes, which are reconstructed from an endoscope image, onto the field of view. The feature of the proposed system is that it reconstructs 3D shapes from the images captured by the endoscope and superimposes them onto the real views. As a result, the superimposed view allows the doctor to keep operating the endoscope while observing the patient's internal body with additional information. The proposed system is composed of the reconstruction module and the display module. The reconstruction module is for acquiring 3D shapes based on an active stereo method. In particular, we propose a novel projection pattern that can reconstruct wide areas of the endoscopic view. The display module shows the 3D shape obtained by the reconstructed module, superimposing on the field of view. In the experiments, we show that it is possible to perform a wide range of dense 3D reconstructions using the new projection patterns. In addition, we confirmed the usefulness of the AR system by interviewing medical doctors.

Active Stereo Method for 3D Endoscopes using Deep-layer GCN and Graph Representation with Proximity Information.

Techniques for 3D endoscopic systems have been widely studied for various reasons. Among them, active stereo based systems, in which structured-light patterns are projected to surfaces and endoscopic images of the pattern are analyzed to produce 3D depth images, are promising, because of robustness and simple system configurations. For those systems, finding correspondences between a projected pattern and an original pattern is an open problem. Recently, correspondence estimation by graph neural networks (GCN) using graph-based representation of the patterns were proposed for 3D endoscopic systems. One severe problem of the approach is that the graph matching by GCN is largely affected by the stability of the graph construction process using the detected patterns of a captured image. If the detected pattern is fragmented into small pieces, graph matching may fail and 3D shapes cannot be retrieved. In this paper, we propose a solution for those problems by applying deep-layered GCN and extended graph representations of the patterns, where proximity information is added. Experiments show that the proposed method outperformed the previous method in accuracies for correspondence matching for 3D reconstruction.