Demonstration of different visual-inertial odometry methods (IMAGE)
Caption
Demonstration of different visual-inertial odometry methods: (a) traditional VIO methods, which rely on handcrafted features and geometry-based optimization; (b) existing deep learning-based methods, which extract features and fuse multi-modal information using deep neural networks; (c) our proposed VIO method, which employs a modal interaction and selection module to enhance the robustness and accuracy of visual-inertial localization.
Credit
Changjun Gu
Usage Restrictions
Credit must be given to the creator. Only noncommercial uses of the work are permitted. No derivatives or adaptations of the work are permitted.
License
CC BY-NC-ND