Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

TartanVO: A Generalizable Learning-based VO

About

We present the first learning-based visual odometry (VO) model, which generalizes to multiple datasets and real-world scenarios and outperforms geometry-based methods in challenging scenes. We achieve this by leveraging the SLAM dataset TartanAir, which provides a large amount of diverse synthetic data in challenging environments. Furthermore, to make our VO model generalize across datasets, we propose an up-to-scale loss function and incorporate the camera intrinsic parameters into the model. Experiments show that a single model, TartanVO, trained only on synthetic data, without any finetuning, can be generalized to real-world datasets such as KITTI and EuRoC, demonstrating significant advantages over the geometry-based methods on challenging trajectories. Our code is available at https://github.com/castacks/tartanvo.

Wenshan Wang, Yaoyu Hu, Sebastian Scherer• 2020

Related benchmarks

TaskDatasetResultRank
Visual-Inertial OdometryEuRoC (All sequences)
MH1 Error0.639
69
SLAMM3DGR
Average Rank8.5
68
Visual OdometryKITTI
KITTI Seq 03 Error2.7
45
Visual OdometryTUM-RGBD
freiburg1/desk2 Error0.122
43
Visual OdometryKITTI Seq. 10
Translational Error (%)6.89
35
Camera pose estimationTUM freiburg1
Rotation Error0.049
34
Visual OdometryKITTI Odometry official (sequences 00-10)
Sequence 10 Error11.9
25
Camera pose estimationSintel 14-sequence
ATE23.8
24
Visual OdometryTartanAir (test)
Error MH0002.12
19
Visual-Inertial OdometryEuRoC MAV
Average Error0.68
19
Showing 10 of 29 rows

Other info

Follow for update