Log In Sign Up

Two Stream Networks for Self-Supervised Ego-Motion Estimation

by   Rares Ambrus, et al.

Learning depth and camera ego-motion from raw unlabeled RGB video streams is seeing exciting progress through self-supervision from strong geometric cues. To leverage not only appearance but also scene geometry, we propose a novel self-supervised two-stream network using RGB and inferred depth information for accurate visual odometry. In addition, we introduce a sparsity-inducing data augmentation policy for ego-motion learning that effectively regularizes the pose network to enable stronger generalization performance. As a result, we show that our proposed two-stream pose network achieves state-of-the-art results among learning-based methods on the KITTI odometry benchmark, and is especially suited for self-supervision at scale. Our experiments on a large-scale urban driving dataset of 1 million frames indicate that the performance of our proposed architecture does indeed scale progressively with more data.


Self-Supervised Ego-Motion Estimation Based on Multi-Layer Fusion of RGB and Inferred Depth

In existing self-supervised depth and ego-motion estimation methods, ego...

Towards Better Generalization: Joint Depth-Pose Learning without PoseNet

In this work, we tackle the essential problem of scale inconsistency for...

Neural Ray Surfaces for Self-Supervised Learning of Depth and Ego-motion

Self-supervised learning has emerged as a powerful tool for depth and eg...

AsyncNeRF: Learning Large-scale Radiance Fields from Asynchronous RGB-D Sequences with Time-Pose Function

Large-scale radiance fields are promising mapping tools for smart transp...

Maximizing Self-supervision from Thermal Image for Effective Self-supervised Learning of Depth and Ego-motion

Recently, self-supervised learning of depth and ego-motion from thermal ...

VLocNet++: Deep Multitask Learning for Semantic Visual Localization and Odometry

Visual localization is one of the fundamental enablers of robot autonomy...

Generalizing to the Open World: Deep Visual Odometry with Online Adaptation

Despite learning-based visual odometry (VO) has shown impressive results...