Semantic Visual Localization

TL;DR

This paper introduces a robust visual localization method that combines 3D geometric and semantic understanding, using a generative descriptor model trained with semantic scene completion, effective under challenging conditions.

Contribution

It presents a novel joint 3D geometric and semantic approach with a generative descriptor model trained on semantic scene completion, improving localization robustness.

Findings

Reliable localization under extreme viewpoint changes

Effective under varying illumination and geometry

Outperforms previous methods on large-scale datasets

Abstract

Robust visual localization under a wide range of viewing conditions is a fundamental problem in computer vision. Handling the difficult cases of this problem is not only very challenging but also of high practical relevance, e.g., in the context of life-long localization for augmented reality or autonomous robots. In this paper, we propose a novel approach based on a joint 3D geometric and semantic understanding of the world, enabling it to succeed under conditions where previous approaches failed. Our method leverages a novel generative model for descriptor learning, trained on semantic scene completion as an auxiliary task. The resulting 3D descriptors are robust to missing observations by encoding high-level 3D geometric and semantic information. Experiments on several challenging large-scale localization datasets demonstrate reliable localization under extreme viewpoint, illumination, and geometry changes.