Gallery

CVC at IoT Solutions World Congress 2019

by

The Computer Vision Center was present at this year’s IoT Solutions World Congress presenting its technology in Computer Vision. A high number of enterprises and companies visited our stand and technological demonstration at the Catalan Pavilion.  

The IoT Solutions World Congress is the largest international event dedicated to IoT (Internet of Things) solutions for industry. This year the event took place from the 29th to the 31st of October at Fira de Barcelona – in the framework of the Barcelona Industry Week – and was attended by more than 16.000 visitors and 350 exhibitors from a wide range of nationalities, with the aim of establishing new partnerships.

The Computer Vision Center highlighted its latest Computer Vision technologies for IoT, with a Smart Market demonstration: a closed circuit that uses new approaches in Deep Learning to detect different packages of grocery items. Without reading a barcode or price, by using computer vision, the system identifies, on site, the item shown from any angle or position, automatically exhibiting the price in the customer’s ticket.

Furthermore, the CVC was also present at the Catalonia AI demo zone with a demonstration on apparent personality detection. This technology has the ability of giving the viewer a profile of apparent personality after just 15 seconds of video. Both demonstrations were highly succesful giving CVC broad recognition within the IOT industry world.

 

Have a look at our IoT Solutions World Congress Moment in Twitter: https://twitter.com/i/moments/1191294755721875456

Neural Networks and personality testing at this year’s Maker Faire 2019

by

Once again, the Computer Vision Center took part in Barcelona’s Maker Faire, which took place on the first weekend of October 2019 at Nau Bostik, an emblematic citizen led cultural center in the neighborhood of La Sagrera (Barcelona). The Faire is visited by more than 5.000 people every year, mainly by families and makers who are enthusiastic of open source projects. The Faire offers diverse perspectives on the great challenges of our progressively complex society. It is a spot that gathers international and local makers, members of industry, universities and research centers who are willing to promote, share and explain their different initiatives and projects.

As in past editions, the CVC showcased some of its Open Source work. The star of the day was our demonstration on apparent personality detection, which uses a thorough analysis of tone, movement and facial microexpressions to predict an apparent personality within a 15-second video. We also presented our Computational Neural Network game, a demonstration aimed to all audiences in which we simplify how a neural network gathers data and takes a decision, which is the basics of how we teach a computer to understand the environment around it by means of cameras.

In the media:

Our participation in this year’s Maker Faire was featured at El País (in Catalan): https://cat.elpais.com/cat/2019/10/04/tecnologia/1570170723_757789.html

CVC Researchers at this year’s ICDAR 2019

by

Several CVC researchers attended the 15th International Conference on Document Analysis and Recognition (ICDAR 2019) that took place in the International Convention Centre in Sydney, Australia, from the 20 to 25th of September. ICDAR is the largest gathering on Document Analysis worldwide, in which CVC normally has an important presence.

The articles presented were the following:

Table Detection in Invoice Documents by Graph Neural Networks, by authors Pau Riba, Anjan Dutta, Lutz Goldmann, Alicia Fornés, Oriol Ramos and Josep Lladós.

Training-Free and Segmentation-Free Word Spotting using Feature Matching and Query Expansion, by authors Ekta Vats, Anders Hast and Alicia Fornés.

Can One Deep Learning Model Learn Script-Independent Multilingual WordSpotting?, by authors Mohammed Al-Rawi, Ernest Valveny and Dimosthenis Karatzas.

Recurrent Comparator with attention models to detect counterfeit documents, by authors Albert Berenguel Centeno, Oriol Ramos Terrades, Josep Lladós i Canet and Cristina Cañero Morales.

ICDAR 2019 RRC on Scene Text Visual Question Answering, by authors Ali Furkan Biten, Rubén Perez Tito, Andres Mafla, Lluis Gomez, Marçal Rusiñol, Minesh Mathew, C.V. Jawahar, Ernest Valveny and Dimosthenis Karatzas

Selective Style Transfer for Text, by authors Raul Gomez, Ali Biten, Dimosthenis Karatzas, Lluis Gomez, Jaume Gibert and Marçal Rusiñol

 

 

Have a look at our ICDAR 2019 Moment in Twitter: https://twitter.com/i/moments/1192061665992433665?s=13

ACMCV 2019 – 6th Annual Catalan Meeting on Computer Vision

by

The 6th Annual Catalan Meeting on Computer Vision (ACMCV2019) took place this year on the 17th of September at the Espai Bital, in Barcelona. It is organized every year by the Computer Vision Center and the coordinators of the Master in Computer Vision of several Catalan universities (UAB, UOC, UPC and UPF). The goals of these anual meetings are to strengthen the Catalan academic and industrial Computer Vision network, to highlight the most relevant research results and to allow students from the Master in Computer Vision to meet and interact with members of the Catalan CV community.

This edition, as always, hosted the master thesis presentations of the students of the Master in Computer Vision, a poster session and two keynote lectures carried out by Dr. Gerard Pons-Moll, from the Max Planck Institute for Informatics, with the title ‘Learning from Digital Humans from Images, Videos and Scans’, and Dr. Agata Lapedriza, from the Universitat Oberta de Catalunya whose talk was focused on the evolution of her research in Computer Vision during the last ten years.

Presentation of the Computer Vision Catalan Alliance

Furthermore, we had the kick-off presentation of the Computer Vision Catalan Alliance. This terriotorial alliance aims to promote and foster collaboration within the Computer Vision ecosystem, joining the expertise of  researchers and entrepreneurs in this discipline who are currently based in Catalonia. Dr. Josep Lladós, CVC’s Director, gave a brief introduction and context of the Alliance. Directly after that, there was a panel discussion on the pilars of the Computer Vision Catalan Ecosystem with actors from both industry and Academia: Dr. Elisenda Bou (Vilynx), Mr. Oscar Sala (MWC Barcelona), Dr. Marco Bressan (Satellogic) and Dr. Xavier Giró (IDEAI-UPC). The debate was moderated by Xantal Llavina (TV3 journalist) and closed by the Director of Innovation and Digital Economy of the Catalan Government, Mr. Daniel Marco.

The program and more details about the event can all be found here: http://acmcv.cat/

 

Have a look at our ACMCV 2019 Moment in Twitter: https://twitter.com/i/moments/1192056946683187200?s=13

NeuroBiT team at ECVP 2019

by

CVC’s NeuroBit team presented a talk on colour induction and three posters on computational modelling of V1, visual saliency stimulus generation and symmetry detection at the 42nd edition of the European Conference on Visual Perception (ECVP 2019), which took place in Leuven, Belgium, from August 25th – 29th, 2019.

Our researchers presented the following:

Modelling symmetry perception with banks of quadrature convolutional Gabor kernels by authors Alejandro Párraga, Xavier Otazu and Arash Akbarinia.

A multilayer computational model of the parvocellular pathway in V1 by authors Xim Cerdà-Company, Xavier Otazu and Olivier Panacchio.

Generating Synthetic Images for Visual Attention Modeling, by authors David Berga, Xosé R. Fdez-Vidal, Xavier Otazu, Xosé M. Pardo and Victor Leborán.

Is color assimilation only due to a luminance-chromatic interaction? by authors Xavier Otazu and Xim Cerdà-Company.

CVCR&D 2019

by

On July 25th, we celebrated CVC’s 14th Workshop on Computer Vision Trends and Challenges (CVCR&D2019), the annual half-day workshop in which PhD students at CVC present their current research, as well as their  future lines of work. This year’s Organising Committee has been formed by Dr. Dena Bazazian and Dr. Jose Antonio Iglesias.

The workshop comprised four different oral sessions with 5 minute presentations dedicated to different Computer Vision areas: (1) Machine Vision, Appearance Analysis and Applications, (2) From Music Scores to Hand Written Texts: Robust Reading and Document Analysis, (3) Autonomous Driving, Action Recognition and Siamese Architectures and (4) Learning to Learn.

Click here to view the CVCR&D 2019 program

Furthermore, an informal poster session was held during both coffee breaks. This Conference, the center’s star internal event, is an excellent chance for PhD students to share their work with the CVC community, giving them the opportunity to receive valuable feedback.

 

Artur Mas, 129th President of the Catalan Government, visits CVC

by

The Computer Vision Center received a most honouring visit from a Catalan ex president, with the opportunity to showcase its technology and prototypes to a set of institutional actors and legislators, key to the regional government of Catalonia.

Last Monday 8th of July, the Right Honourable Mr. Artur Mas, 129th President of the Catalan Government, visited the Computer Vision Center, along with a set of distinguished legislators from the regional government. The visit had the aim to know, first-hand, about our research and development activities and counted with the presence of The Honorable Mr. Jordi Puigneró, Conseller of Digital Policies and Public Administration of Catalonia, Mr. Daniel Marco, Director of Innovation and Digital Economy of the Catalan Government, Mrs. Joana Barbany, General Director of Digital Society in the Catalan Government, The Excellent and Magnificent Dr. Margarita Arboix, Rector of the UAB, and the Excellent Dr. Javier Lafuente, UAB’s Vice Rector for Innovation and Strategic Projects.  

During the visit, the President met with Dr. Josep Lladós, CVC Director, Dr. Dimosthenis Karatzas and Dr. Fernando Vilariño, CVC Associate Directors and Dr. Juan José Villanueva, CVC founder and former director. It started with an official greeting and reception followed by an institutional presentation and technology demonstrations by part of several CVC researchers. 

NeuroBiT team at the Iberian Conference on Perception 2019

by

CVC’s NeuroComputation and Biological Vision Team (NeuroBiT) gave three talks and presented one poster at the 8th Iberian Conference on Perception, which was held from the 20th to the 22nd of June in San Lorenzo de El Escorial, Madrid, Spain.

The conference is focused on Perception, emphasizing different aspects like: Motion Perception, Spatial Vision,Stereopsis, Colour Perception, Perception and Action, Attention and Cognition, Auditory Perception, Multisensory Integration and Reading/Speech Perception. Dr. Xavier Otazu was also the organizer of the conference’s symposium Computational Perception.

The papers presented in this Conference were the following:

Poster:

C. Alejandro Parraga, Xavier Otazu, Arash Akbarinia: Modelling symmetry perception with banks of quadrature convolutional Gabor kernels.

Talks:

D. Berga: Computational modeling of visual attention: What do we know from physiology and psychophysics?

D. Berga: Measuring bottom-up visual attention in eye tracking experimentation with synthetic images

Dr. X. Otazu: No chromatic-chromatic interaction in colour assimilation.

5 CVPR’s from CVC this year

by

As every year, the CVC has been present at the most important gathering in the Computer Vision Science Community, the annual Computer Vision and Pattern Recognition conference (CVPR). In this edition, CVC researchers have presented a total of 5 papers. The conference was held from the 16th to the 20th of June in Long Beach, California.

The articles presented were the following:

Doodle to Search: Practical Zero-Shot Sketch-based Image Retrieval , by authors Sounak Dey, Pau Riba, Anjan Dutta, Josep Lladós and Yi-Zhe Song.

Learning Metrics from Teachers: Compact Networks for Image Embedding, by authors Lu Yu, Vacit Oguz Yazici, Xialei Liu, Joost van de Weijer, Yongmei Cheng and Arnau Ramisa

Semantically Tied Paired Cycle Consistency for Zero-Shot Sketch-based Image Retrieval, by authors Anjan Dutta and Zeynep Akata

What does it mean to learn in deep networks? And, how does on detect adversarial attacks?, by authors Ciprian A. Corneanu, Meysam Madadi, Sergio Escalera and Aleix M. Martinez.

Good News, Everyone! Context driven entity-aware captioning for news images, by authors Ali Furkan Biten, Lluis Gomez, Marçal Rusiñol and Dimosthenis Karatzas

Read more

CVC Open Day for students 2019

by

Last Friday 31st of May, CVC celebrated its Open Day in a two turn visit. Students from the Autonomous University of Barcelona and professionals from several local companies visited our institution in order to know, first hand, the research that is being developed in the area of Computer Vision, as well as the different job and training opportunities offered. Moreover, they were able to explore our facilities and test some of our latest demonstrations.