DeepFactors: Real-Time Probabilistic Dense Monocular SLAM
This work addresses the need for robust and efficient monocular SLAM for interactive robotics and augmented reality, representing an incremental advancement by integrating existing approaches into a unified probabilistic system.
The paper tackles the problem of estimating dense geometry and camera motion from monocular images by unifying sparse, dense, and learned methods in a probabilistic SLAM framework, achieving real-time performance with improved trajectory and depth reconstruction on real-world sequences.
The ability to estimate rich geometry and camera motion from monocular imagery is fundamental to future interactive robotics and augmented reality applications. Different approaches have been proposed that vary in scene geometry representation (sparse landmarks, dense maps), the consistency metric used for optimising the multi-view problem, and the use of learned priors. We present a SLAM system that unifies these methods in a probabilistic framework while still maintaining real-time performance. This is achieved through the use of a learned compact depth map representation and reformulating three different types of errors: photometric, reprojection and geometric, which we make use of within standard factor graph software. We evaluate our system on trajectory estimation and depth reconstruction on real-world sequences and present various examples of estimated dense geometry.