We have been working on a number of projects concerning the problem of real-time 3D reconstruction using depth cameras with applications in medical augmented reality. The videos below show a couple of ideas and systems that we have developed. Further work may be carried out on multi-camera reconstruction for 3D recording in the operating room. Immediate applications of such systems are surgical training and medical education. The key here is to bring in machine learning based image analysis to extract clinically relevant information that physicians may need during intervention.
2015 Imperial Festival Demo Teaser.
The video shows an early prototype of our augmented reality system which makes use of the KinectFusion algorithm for real-time surface reconstruction.
This video demonstrates potential use for medical augmented reality.
This video demonstrates our work on real-time camera relocalization.
Dense Planar SLAM presented at ISMAR 2014 by Renato Salas-Moreno.
ElasticFusion presented at RSS 2015 by Thomas Whelan.