/tandem

TANDEM: Tracking and Dense Mapping
in Real-time using Deep Multi-view Stereo

Lukas Koestler1*    Nan Yang1,2*    Niclas Zeller2,3    Daniel Cremers1,2

1Technical University of Munich    2Artisense
3Karlsruhe University of Applied Sciences

Conference on Robot Learning (CoRL) 2021

Abstract

In this paper, we present TANDEM a real-time monocular tracking and dense mapping framework. For pose estimation, TANDEM performs photometric bundle adjustment based on a sliding window of keyframes. To increase the robustness, we propose a novel tracking front-end that performs dense direct image alignment using depth maps rendered from a global model that is built incrementally from dense depth predictions. To predict the dense depth maps, we propose Cascade View-Aggregation MVSNet (CVA-MVSNet) that utilizes the entire active keyframe window by hierarchically constructing 3D cost volumes with adaptive view aggregation to balance the different stereo baselines between the keyframes. Finally, the predicted depth maps are fused into a consistent global map represented as a truncated signed distance function (TSDF) voxel grid. Our experimental results show that TANDEM outperforms other state-of-the-art traditional and learning-based monocular visual odometry (VO) methods in terms of camera tracking. Moreover, TANDEM shows state-of-the-art real-time 3D reconstruction performance.

Poster

For more details, please see:

  • OpenReview for the full paper and supplementary material.

  • Webpage: go.vision.in.tum.de/tandem for further information.

  • Demo Video: YouTube for a live presentation of TANDEM.

  • Code and data coming soon. We are currently preparing a live demo for CoRL and will update this repository afterwards. Thank you for your patience.