Information-Theoretic Odometry Learning

Sen Zhang, Jing Zhang, Dacheng Tao

Keywords: Odometry Learning, Information Bottleneck, Generalization Bound

Abstract: In this paper, we propose a unified information-theoretic framework for odometry learning, a crucial component of many robotics and vision tasks such as navigation and virtual reality where 6-DOF poses are required in real time. We formulate this problem as optimizing a variational information bottleneck objective function, which eliminates pose-irrelevant information from the latent representation. The proposed framework provides an elegant tool for performance evaluation and understanding in information-theoretic language. Specifically, we bound the generalization errors of the deep information bottleneck framework and the predictability of the latent representation. These provide not only a performance guarantee but also practical guidance for model design, sample collection, and sensor selection. Furthermore, the stochastic latent representation provides a natural uncertainty measure without the needs for extra structures or computations. Experiments on two well-known odometry datasets demonstrate the effectiveness of our method.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

One-sentence Summary: An information-bottleneck motivated odometry learning framework with theoretical performance guarantee w.r.t. the generalization error.

Reviewed Version (pdf): https://openreview.net/references/pdf?id=UaIZVkRpCR