MV-dVRK: A Multi-Viewpoint Benchmark for Spatial Surgical Perception
Updated
Updated · arxiv.org · Sep 2
MV-dVRK: A Multi-Viewpoint Benchmark for Spatial Surgical Perception
1 articles · Updated · arxiv.org · Sep 2
Summary
Researchers have introduced MV-dVRK, the first ex-vivo surgical dataset featuring multiple exposure-synchronized stereo viewpoints with validated surface geometry and camera poses.
MV-dVRK enables systematic benchmarking of 3D reconstruction methods in minimally invasive surgery, comparing monocular, stereo, multi-stereo, and multi-view techniques as viewpoints increase.
This advance provides a new standard for evaluating spatial perception in surgical robotics and supports future research into dynamic, multi-viewpoint surgical 3D perception.