Real-time 3D Pose Estimation with a Monocular Camera Using Deep Learning and Object Priors On an Autonomous Racecar.
Ankit DhallPublished in: CoRR (2018)
Keyphrases
- monocular camera
- deep learning
- real time
- ground plane
- visual odometry
- multibody
- autonomous navigation
- inertial sensors
- articulated objects
- rigid objects
- unsupervised learning
- machine learning
- position and orientation
- multiple objects
- feature tracking
- weakly supervised
- sensor fusion
- d objects
- long range
- point correspondences
- multiple views
- object tracking
- human body
- image sequences