Direct-PoseNet: Absolute Pose Regression with Photometric Consistency

Abstract

We present a relocalization pipeline, which combines an absolute pose regression (APR) network with a novel view synthesis based direct matching module, offering superior accuracy while maintaining low inference time. Our contribution is twofold: i) we design a direct matching module that supplies a photometric supervision signal to refine the pose regression network via differentiable rendering; ii) we show that our method can easily cope with additional unlabeled data without the need for external supervision such as traditional visual odometry or pose graph optimization. As a result, our method achieves state-of-the-art performance among all other single-image APR methods on the 7-Scenes benchmark and the LLFF dataset.