3D Vision - NAVER LABS Europe


In a connected world of people, robots and self-driving vehicles, we naturally need to have a good understanding of the 3D world we live in.



Accepted paper at CVPR:
Large-scale localization datasets in crowded indoor spaces. Comes with release of world’s biggest indoor localization dataset and a new version of the unified data format kapture!


New Image Retrieval for Visual Localization Benchmark online.

New Kapture toolbox online.

4 outstanding reviewer awards: M. Humenberger (CVPR, 3DV), J. Revaud (CVPR, NeurIPS)

2 papers accepted at 3DV 2020:

2 papers accepted at NeurIPS 2020:

IJCV: Volume Sweeping: Learning Photoconsistency for Multi-View Shape Reconstruction

2nd place in Long-Term Visual Localization under Changing Conditions held at ECCV20 (paper).

Kapture: Release of a unified data format and processing pipeline for structure from motion and visual localization (paper).

2 papers ECCV 2020:

IROS 2020 paper: Self-Supervised Attention Learning for Depth and Ego-motion Estimation

IEEE Access journal: The IPIN 2019 Indoor Localisation Competition – Description and Results by F. Potorti, B. Chidlovskii, L. Antsfeld, et al.

1st, 2nd and 4th in CVPR Long Term Visual Localization Challenge.

New version of popular synthetic dataset Virtual KITTI available.

Related Content


NAVER LABS Europe 3D vision

In a connected world of people, robots and self-driving vehicles, we naturally need to have a good understanding of the 3D world we live in.

Results in the tasks related to this understanding such as 3D reconstruction, mapping and visual localization have been getting better and better. In reconstructing the geometry of the world as accurately as possible, it’s common practice to use sensors such as LIDAR, radar and, of course, cameras. This is because geometry is pretty well understood and one needs to ‘measure the world’ for many applications. However, progress in only using geometry to solve 3D vision tasks has been declining and the methods that exist today are not sufficiently robust for everyday situations such as changing environments and weather conditions.

One reason for this lack of robustness is that not everything can be measured or described in a way a computer can reliably detect it. Furthermore, even if a scene were to be perfectly reconstructed, there’s no guarantee that a computer would understand, analyse and interpret it correctly.

A popular strategy of the computer vision community to overcome these problems, is to use machine learning techniques rather than hand-crafted approaches and their success has proven it to be a good choice. There have been some outstanding results in topics such as image categorization, image retrieval and object detection.

However, geometric properties constitute a significant part of the world and we believe they should not be neglected entirely in favour of learning. Our strategy therefore combines both approaches.

We want to learn what we cannot measure.

The research focus of the 3D Vision team lies on the design of methods which combine geometry and learning-based approaches to solve specific real-world challenges such as visual localization, camera pose estimation and 3D reconstruction. Examples for our target applications are robot navigation, indoor mapping, augmented reality (AR) and, more generally speaking, systems which enable ambient intelligence in day to day life.

Localization Datasets in Crowded Indoor Spaces
NAVER LABS has made available five new indoor datasets for large scale visual localization in crowded public spaces. Blog article by Donghwan Lee, Soohyun Ryu, Suyong Yeon, Yonghan Lee, Deokhwa Kim, Cheolho Han, Yohann Cabon, Philippe Weinzaepfel, Nicolas Guerin, Gabriela Csurka Khedari and Martin Humenberger
Methods for visual localization blog image
This article gives an overview of current state-of-the-art methods and their advantages and drawbacks. Blog article by Martin Humenberger, Gabriela Csurka Khedari, Nicolas Guerin and Boris Chidlovskii
Open source release of the structure from motion and visual localization data format kapture. Blog article by Martin Humenberger
One Method One Pipeline Blog Image
Using the KAPTURE pipeline and robust R2D2 method, we ranked 1st, 2nd and 4th in the VisLocOdomMap workshop challenge. Blog article by Martin Humenberger

3D Vision team:

Leonid Antsfeld
Romain Brégier
Yohann Cabon
Dorian Goepp
Nicolas Guerin
Martin Humenberger
Vincent Leroy
Gianluca Monaci
Julien Morat
Maxime Pietrantoni
PhD candidate
Hervé Poirier
Philippe Rerole
Jérome Revaud
Assem Sadek
PhD candidate