DPC-Net: Deep Pose Correction for Visual Localization
This work addresses the challenge of enhancing efficiency and accuracy in visual odometry for robotics and autonomous systems, though it is incremental as it builds on existing estimators.
The paper tackles the problem of improving visual localization accuracy by using a deep network to correct errors in classical geometric estimators, achieving performance comparable to computationally-intensive dense estimators on the KITTI odometry dataset.
We present a novel method to fuse the power of deep networks with the computational efficiency of geometric and probabilistic localization algorithms. In contrast to other methods that completely replace a classical visual estimator with a deep network, we propose an approach that uses a convolutional neural network to learn difficult-to-model corrections to the estimator from ground-truth training data. To this end, we derive a novel loss function for learning SE(3) corrections based on a matrix Lie groups approach, with a natural formulation for balancing translation and rotation errors. We use this loss to train a Deep Pose Correction network (DPC-Net) that predicts corrections for a particular estimator, sensor and environment. Using the KITTI odometry dataset, we demonstrate significant improvements to the accuracy of a computationally-efficient sparse stereo visual odometry pipeline, that render it as accurate as a modern computationally-intensive dense estimator. Further, we show how DPC-Net can be used to mitigate the effect of poorly calibrated lens distortion parameters.