FullResults_GoodFeature

FullResults_GoodFeature copied to clipboard

Devices

• 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

Benchmark

• 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

VO/VSLAM/VINS Systems Assessed

• 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.

References

@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}
}	

Contact information

• Yipu Zhao [email protected]
• Patricio A. Vela [email protected]