My current research deals with the following questions:
1) How to generate photorealistic synthetic data to overcome data imbalance and lack of labeled data in surgical scenes for 3D vision tasks?
2) How to enable (surgical) robots to spatially understand their environment/surroundings?