- Desktop (passmark 2584, 95W) https://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i7-7700K+%40+4.20GHz&id=2874
- X200 (passmark 905, 17W) https://www.cpubenchmark.net/cpu.php?cpu=Intel+Pentium+2117U+%40+1.80GHz&id=1872
- Jetson TX2 (-, 7.5W) https://www.phoronix.com/scan.php?page=article&item=march-2017-arm&num=2
- Euclid (passmark 552, 4W) https://www.cpubenchmark.net/cpu.php?cpu=Intel+Atom+x7-Z8700+%40+1.60GHz&id=2506
- EuRoC Mono & Stereo https://projects.asl.ethz.ch/datasets/doku.php?id=kmavvisualinertialdatasets
- TUM VI Mono & Stereo https://vision.in.tum.de/data/datasets/visual-inertial-dataset
- TUM RGBD Mono https://vision.in.tum.de/data/datasets/rgbd-dataset
- ICL NUIM Mono https://www.doc.ic.ac.uk/~ahanda/VaFRIC/iclnuim.html
- KITTI Stereo http://www.cvlibs.net/datasets/kitti/eval_odometry.php
- ORB-SLAM Mono with/without Good Feature https://github.com/ivalab/GF_ORB_SLAM
- ORB-SLAM Stereo with/without Good Feature https://github.com/ivalab/gf_orb_slam2
- SVO Mono & Stereo https://github.com/YipuZhao/rpg_svo
- DSO Mono https://github.com/YipuZhao/DSO
- DSO Stereo [results downloaded from https://vision.in.tum.de/research/vslam/stereo-dso]
- ROVIO Mono https://github.com/YipuZhao/rovio
- VINS-Mono https://github.com/YipuZhao/VINS-Mono
- MSCKF Stereo https://github.com/YipuZhao/msckf_vio
- OKVIS Stereo https://github.com/YipuZhao/okvis
Compared to the official repos, the baseline methods in my repo are with explicit logging on pose tracking and time cost.
@article{zhao2019good,
title={Good Feature Matching: Towards Accurate, Robust VO/VSLAM with Low Latency},
author={Zhao, Yipu and Vela, Patricio A.},
journal={submitted to IEEE Transactions on Robotics},
year={2019}
}
- Yipu Zhao yipu.zhao@gatech.edu
- Patricio A. Vela pvela@gatech.edu