NAVER at ECCV 2020 - Naver Labs Europe

NAVER at ECCV 2020

23rd-28th August 2020

NAVER, LINE and NAVER LABS are ECCV2020 Silver sponsors 

NAVER virtual ECCV booth

NAVER is Korea’s premier internet company and a global leader in online services like NAVER search, LINE messaging and WEBTOON. NAVER invests over a quarter of revenue in R&D and, through advanced technology, is seamlessly connecting the physical and digital worlds. Its AI and Robotics research in Asia and Europe is fundamental to creating this future.

CLOVA AI Research is the team responsible for advanced and fundamental AI technologies based on machine learning and deep learning that enable the global NAVER and LINE AI platform (CLOVA) to be smarter. Computer vision, natural language processing, recommendation, and pattern recognition technologies are upgraded by establishing new models and investigating optimization methods suitable for existing models.  Results are disseminated on the NAVER github site and in conference publications such as CVPR, ECCV, ICCV, NeurIPS, ICML, Interspeech, EMNLP…

NAVER LABS is an R&D subsidiary of NAVER responsible for future technology. Its world class researchers in Korea and Europe create new connections between people, machines, spaces and information by advancing technology in AI, robotics, autonomous driving, 3D/HD mapping and AR.

************************************************ ECCV 2020 NAVER NEWS ************************************************

Florent Perronnin

Congratulations to Florent Perronnin,  NAVER LABS Europe, Jorge Sanchez National University of Cordoba & Thomas Mensink, University of Amsterdam for winning the Koenderink 10 year test of time award at ECCV 2020 for their 2010 paper ‘Improving the Fisher kernel for large-scale image classification’ !  The award was announced in the opening ceremony of the conference Monday August 24th. 


NAVER LABS Europe came 2nd in the Long-Term Visual Localization Challenge in Autonomous Vehicles!





A live discussion: Wednesday, 26th August 2020
11:30-12:15pm (UTC+1)  12:30-13:15  CET  19:30-20:15 KST

NAVER is presenting 9 papers at ECCV as well as giving tutorials and organising workshops. In this live session, you’ll get a brief overview of relevant R&D activities and career opportunities at NAVER, Korea’s no.1 Internet portal and leading global content services provider. Leaders from NAVER LABS research in Korea and Europe and the AI division ‘Clova AI’ will present current work in computer vision, machine learning, AR, AI and Robotics followed by a Q&A.

Naila Murray

Naila Murray
Director of Science, NAVER LABS Europe. Princeton University and Universitat Autonoma de Barcelona & CVC Barcelona. Previously head of computer vision research at NAVER LABS Europe. Research interests include representation learning and multi-modal search. AC at CVPR 2020, PC for ICLR 2021.


Jongyoon Peck
Executive Officer and Head of Autonomous Driving at NAVER LABS Korea, Member of the Korean Presidential Committee on 4th Ind. Revolution, Previously Samsung Techwin,  Stanford University and Seoul National University.


Jung-Woo Ha
Executive Officer and Head of Research at Clova AI. Clova is a general-purpose AI assistance platform developed by NAVER and LINE. PhD in machine learning and AI from Seoul National University.

NAVER LABS Europe is supporting the Visual Domain Adaptation Challenge (VisDA-2020) and the TASK-CV 2020, Robust Vision Challenge, RCV 2020 and giving a tutorial on Domain Adaptation for Visual Applications.

Publications, Workshops and Tutorials at ECCV 2020


24th August 06:00 and 14:00 #1014
Learning Visual Representations with Caption Annotations [supplementary material]
Bülent Sariyildiz, Julien Perez, Diane Larlus

25th August 14:00 and 20:00 #2074
ReAD: Reciprocal Attention Discriminator for Image-to-Video Re-Identification  [supplementary material]
Minho Shim, Hsuan-I Ho, Jinhyung Kim, Dongyoon Wee. 

25th August 14:00 and 20:00 #2241
Character Region Attention For Text Spotting
Youngmin Baek, Seung Shin, Jeonghun Baek, Sungrae Park, JunyeopLee, Daehyun Nam, Hwalsuk Lee

25th August 14:00 and 26th August 00:00 #2360
Measuring Generalisation to Unseen Viewpoints, Articulations, Shapes and Objects for 3D Hand Pose Estimation under Hand-Object Interaction [supplementary material]
Anil Armagan, Philippe Weinzaepfel, Romain Bregier, Gregory Rogez et al.  

26th August 06:00 and 14:00 #3214
DOPE: Distillation Of Part Experts for whole-body 3D pose estimation in the wild  [Blog] 
Philippe Weinzaepfel, Romain Brégier, Hadrien Combaluzier, Vincent Leroy, Gregory Rogez

26th August 22:00 and 27th August 08:00 #3241
BSL-1K: Scaling up co-articulated sign recognition using mouthing cues  [supplementary material]
Samuel Albanie, Gul Varol, Liliane Momeni, Triantafyllos Afouras, Joon Son Chung, Neil Fox, Andrew Zisserman

27th August 14:00 and 20:00 #4137
Few-shot Compositional Font Generation with Dual Memory.  [supplementary material]
Junbum Cha, Sanghyuk Chun, Gayoung Lee, Bado Lee, Seonghyeon Kim, Hwalsuk Lee

27th August 14:00 and 28th August 00:00 #4335
Learning to Generate Grounded Visual Captions without Localization Supervision  [supplementary material]
Chih-Yao Ma, Yannis Kalantidis, Ghassan AlRegib, Peter Vajda, Marcus Rohrbach, Zsolt Kira 

27th August 22:00 and 28th August 08:00 #4330
Self-supervised learning of audio-visual objects from video  [supplementary material]
Triantafyllos Afouras, Andrew Owens, Joon Son Chung, Andrew Zisserman


Workshop: TASK-CV
Date: Sunday 23rd August 2020
Organisers: Tatiana Tommasi (Politecnico di Torino, Italy), Antonio M. Lopez (CVC & UAB, Spain), David Vazquez (Element AI, Canada), Gabriela Csurka (NAVER LABS Europe, France), Kate Saenko (Boston University, USA), Liang Zheng (Australian National University, Australia), Xingchao Peng (Boston University, USA), Weijian Deng (Australian National University, Australia)

NAVER LABS Europe is supporting the best paper award at TASK-CV 2020.

Workshop: Women in Computer Vision (WiCV)
Date: Sunday 23rd August, 10pm (UTC+1)
Diane Larlus, Principal Scientist, mentor at the mentoring dinner

Workshop: Long-Term Visual Localization
Date: Friday 28th August 2020
ChallengeNAVER LABS Europe 2nd place in Autonomous Vehicle Challenge: Martin Humenberger, Yohann Cabon, Nicolas Guerin, Julien Morat, Jérôme Revaud, Philippe Rerole, Noe ́ Pion, Cesar de Souza, Gabriela Csurka, Robust Image Retrieval-based Visual Localization using Kapture.

All datasets for this challenge were provided by NAVER LABS Europe’ in the kapture format,  a data format and tool set for facilitating the integration of various datasets within visual localization processing environments.
More details on can be found on the kapture website.
Code, specifications, and documentation are on GitHub.

NAVER LABS Europe is a sponsor of the Long-Term Visual Localization Workshop.

Tutorial: Weakly supervised learning in computer vision
Date: Friday 28th August, 9am and 5pm (UTC+1) 
Speakers: Haken Bilen, University of Edinburgh, Rodrigo Benenson, Google and Seong Joon Oh, NAVER

Workshop: Instance-Level Recognition Workshop
Date: Friday 28th August, 10pm (UTC+1)
Invited talk: Diane Larlus, Principal Scientist, “From Instance-Level to Semantic Image Retrieval”

delegate flyer ECCV2020

NAVER and NAVER LABS publish annually at CVPR, ECCV/ICCV, NeurIPS, ICML, ACL, EMNLP,  SIGIR, RecSys and many other conferences.

More publications are listed here:

NAVER was recognised as a top employer and company university students would like to work for in South Korea for 3 consecutive years (2016 – 2019)

team photo

Would you like to join one of the world’s most innovative companies and
have impact on the lives of millions of people?


Diversity is the reason NAVER came into existence in 1999. The need to provide alternatives is a fundamental core value for a healthy society.

mind sharing image

We value different ways of thinking about the world and different perceptions of the world.

button transparent

We try to create an inclusive workplace where respect reigns. A place where everyone can be themselves.

NAVER headquater
NAVER HQ ‘Green Factory’, Seongnam, Korea
NAVER LABS Europe, Grenoble, France
NAVER LABS Korea view from the kitchen
NAVER LABS, Seongnam-si, Korea (view from the coffee bar)

NAVER was recognised as a top employer and company university students would like to work for in South Korea for 3 consecutive years
(2016 – 2019)

Recent News:

Rethinking the Truly Unsupervised Image-to-Image Translation
Extracting road layout information from aerial images through deep learning technology