Visual Odometry for Self-Driving with Multihypothesis and Network Prediction

Abstract

Robustness in visual odometry (VO) systems is critical, as it determines reliable performance in various scenarios and challenging environments. Especially with the development of data-driven technology, such as deep learning, the combination of data-driven VO and traditional model-based VO has achieved accurate tracking performance. However, the existence of local optimums in the model-based cost function still limits the robustness. In this study, we introduce a novel framework with a particle filter (PF) in the optimization process, where the PF is constructed by deep neural network (DNN) prediction. We propose constructing the PF by motion prediction classification and its uncertainty based on the characteristic of on-road driving motion. At the same time, an interval DNN prediction strategy is introduced to improve the real-Time performance. Experimental results show that our framework obtains better tracking accuracy and robustness than the existing works, while the time consumption is maintained.