Transformer-Based Model for Monocular Visual Odometry: A Video Understanding Approach
This addresses the problem of camera pose estimation for mobile robots and autonomous vehicles, offering a generalizable deep learning approach that reduces engineering effort compared to traditional geometric methods, though it is incremental as it builds on existing Transformer architectures.
The authors tackled monocular visual odometry by framing it as a video understanding task, using a Transformer-based model called TSformer-VO to estimate camera pose from clips, achieving competitive state-of-the-art performance on the KITTI dataset and outperforming the DeepVO implementation.
Estimating the camera's pose given images from a single camera is a traditional task in mobile robots and autonomous vehicles. This problem is called monocular visual odometry and often relies on geometric approaches that require considerable engineering effort for a specific scenario. Deep learning methods have been shown to be generalizable after proper training and with a large amount of available data. Transformer-based architectures have dominated the state-of-the-art in natural language processing and computer vision tasks, such as image and video understanding. In this work, we deal with the monocular visual odometry as a video understanding task to estimate the 6 degrees of freedom of a camera's pose. We contribute by presenting the TSformer-VO model based on spatio-temporal self-attention mechanisms to extract features from clips and estimate the motions in an end-to-end manner. Our approach achieved competitive state-of-the-art performance compared with geometry-based and deep learning-based methods on the KITTI visual odometry dataset, outperforming the DeepVO implementation highly accepted in the visual odometry community. The code is publicly available at https://github.com/aofrancani/TSformer-VO.