DeepTAM: Deep Tracking and Mapping
This work addresses the challenge of accurate 6 DOF tracking and mapping for computer vision applications, offering a competitive alternative to RGB-D and classic methods.
The authors tackled the problem of dense camera tracking and depth map estimation by developing a fully learned system that simplifies pose estimation and improves accuracy through hypothesis generation, achieving state-of-the-art results with few images and robustness to noise.
We present a system for keyframe-based dense camera tracking and depth map estimation that is entirely learned. For tracking, we estimate small pose increments between the current camera image and a synthetic viewpoint. This significantly simplifies the learning problem and alleviates the dataset bias for camera motions. Further, we show that generating a large number of pose hypotheses leads to more accurate predictions. For mapping, we accumulate information in a cost volume centered at the current depth estimate. The mapping network then combines the cost volume and the keyframe image to update the depth prediction, thereby effectively making use of depth measurements and image-based priors. Our approach yields state-of-the-art results with few images and is robust with respect to noisy camera poses. We demonstrate that the performance of our 6 DOF tracking competes with RGB-D tracking algorithms. We compare favorably against strong classic and deep learning powered dense depth algorithms.