# Visual Localization Using Sparse Semantic 3D Map

arXiv: https://arxiv.org/abs/1904.03803 · HTML: https://mlsanity.com/paper/1904.03803/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Apr 2019.

## Built on by (most cited first, 9 of 9)

- [SFD2: Semantic-guided Feature Detection and Description](https://mlsanity.com/paper/2304.14845/) (2023)
- [City-scale Scene Change Detection using Point Clouds](https://mlsanity.com/paper/2103.14314/) (2021)
- [Semantic Image Alignment for Vehicle Localization](https://mlsanity.com/paper/2110.04162/) (2021)
- [Weatherproofing Retrieval for Localization with Generative AI and Geometric Consistency](https://mlsanity.com/paper/2402.09237/) (2024)
- [Deep auxiliary learning for visual localization using colorization task](https://mlsanity.com/paper/2107.00222/) (2021)
- [Towards Robotic Knee Arthroscopy: Multi-Scale Network for Tissue-Tool Segmentation](https://mlsanity.com/paper/2110.02657/) (2021)
- [Crowd Source Scene Change Detection and Local Map Update](https://mlsanity.com/paper/2203.05205/) (2022)
- [PRAM: Place Recognition Anywhere Model for Efficient Visual Localization](https://mlsanity.com/paper/2404.07785/) (2024)
- [Map-Free Visual Relocalization Enhanced by Instance Knowledge and Depth Knowledge](https://mlsanity.com/paper/2408.13085/) (2024)

## Abstract (arXiv)

Accurate and robust visual localization under a wide range of viewing condition variations including season and illumination changes, as well as weather and day-night variations, is the key component for many computer vision and robotics applications. Under these conditions, most traditional methods would fail to locate the camera. In this paper we present a visual localization algorithm that combines structure-based method and image-based method with semantic information. Given semantic information about the query and database images, the retrieved images are scored according to the semantic consistency of the 3D model and the query image. Then the semantic matching score is used as weight for RANSAC's sampling and the pose is solved by a standard PnP solver. Experiments on the challenging long-term visual localization benchmark dataset demonstrate that our method has significant improvement compared with the state-of-the-arts.