Dense RGB-D-Inertial SLAM with Map Deformations
This addresses robustness issues in dense SLAM for robotics or AR/VR applications, but it is incremental as it extends existing sparse inertial fusion to dense methods.
The authors tackled the problem of dense visual SLAM lacking robustness in tracking, especially with poor initialization, by proposing the first tightly-coupled dense RGB-D-inertial SLAM system. They showed it is more robust to fast motions and low-texture periods than RGB-D-only SLAM, with real-time GPU capability.
While dense visual SLAM methods are capable of estimating dense reconstructions of the environment, they suffer from a lack of robustness in their tracking step, especially when the optimisation is poorly initialised. Sparse visual SLAM systems have attained high levels of accuracy and robustness through the inclusion of inertial measurements in a tightly-coupled fusion. Inspired by this performance, we propose the first tightly-coupled dense RGB-D-inertial SLAM system. Our system has real-time capability while running on a GPU. It jointly optimises for the camera pose, velocity, IMU biases and gravity direction while building up a globally consistent, fully dense surfel-based 3D reconstruction of the environment. Through a series of experiments on both synthetic and real world datasets, we show that our dense visual-inertial SLAM system is more robust to fast motions and periods of low texture and low geometric variation than a related RGB-D-only SLAM system.