Constrained Visual-Inertial Localization With Application And Benchmark in Laparoscopic Surgery
Abstract
We propose a novel method to tackle the visual-inertial localization problem for constrained camera movements. We use residuals from the different modalities to jointly optimize a global cost function. The residuals emerge from IMU measurements, stereoscopic feature points, and constraints on possible solutions in SE(3). In settings where dynamic disturbances are frequent, the residuals reduce the complexity of the problem and make localization feasible. We verify the advantages of our method in a suitable medical use case and produce a dataset capturing a minimally invasive surgery in the abdomen. Our novel clinical dataset MITI is comparable to state-of-the-art evaluation datasets, contains calibration and synchronization.
Overview
a) We propose a visiual odometry algorthm, where we exploit naturally given motion constraints, known in the medical domain as remote center-of-motion (RCM), reducing the solution space to tackle the more challenging image domains, which contain large movements in the image, occlusions, fog, deformable surfaces or difficult illuminations conditions.
b) We introduce a novel dataset suitable for testing SLAM in the minimally-invasive-surgery domain. The dataset contains synchronized stereo images, IMU data, and ground truth poses and is captured during a minimally invasive surgery in the abdomen with a handheld stereo endoscope.