close

Conferences

ConferencesCVC NewsEvents & Community

10 CVC papers presented at ICPR2020

cropped-ICPR2020_virtual-milano

Several CVC researchers attended the 25th International Conference on Pattern Recognition (ICPR’2020), which took place fully virtual on January 10-15th, 2021 due to the pandemic. ICPR is the flagship conference of IAPR the International Association of Pattern Recognition and the premiere conference in pattern recognition, covering computer vision, image, sound, speech, sensor patterns processing and machine intelligence. On this year’s edition, CVC researchers presented 8 papers in total: 3 orals and 5 posters at the main conference and 2 papers at the Conference’s Workshops.

The papers presented at ICPR2020 were the following: 

Named Entity Recognition and Relation Extraction with Graph Neural Networks in Semi Structured Documents, by authors Manuel Carbonell, Pau Riba, Mauricio Villegas, Alicia Fornés and Josep Lladós (oral)

Learning to Rank for Active Learning: A Listwise Approach, by authors Minghan Li, Xialei Liu, Joost Van de Weijer and Bogdan Raducanu (oral)

A Few-Shot Learning Approach for Historical Ciphered Manuscript Recognition, by authors Mohamed Ali Souibgui, Alicia Fornés, Yousri Kessentini and Crina Tudor (Best Student Paper Award) (oral)

Rank-Based Ordinal Classification, by authors Joan Serrat and Idoia Ruiz (poster)

LiNet: A Lightweight Network for Image Super Resolution, by authors Armin Mehri, Parichehr Behjati Ardakani and Angel D. Sappa (poster)

Text Recognition – Real World Data and Where to Find Them, by authors Klára Janouková (CMP-CTU in Prague), Lluis Gomez, Dimosthenis Karatzas and Jiri Matas (poster)

Uncertainty-Aware Data Augmentation for Food Recognition, by authors Eduardo Aguilar, Bhalajo Nagarajan, Rupali Khatun, Marc Bolaños and Petia Radeva (Universitat de Barcelona & CVC) (poster)

Modeling Long-Term Interactions to Enhance Action Recognition, by authors Alejandro Cartas, Petia Radeva and Mariella Dimiccoli (Universitat de Barcelona & CVC) (poster)

Papers presented at the Conference’s Workshops:

Towards Stroke Patients’ Upper-limb Automatic Motor Assessment Using Smartwatches, by authors Asma Bensalah, Jialuo Chen, Alicia Fornés, Cristina Carmona-Duarte, Josep Lladós and Miguel A. Ferrer, presented at the Artificial Intelligence for Healthcare Applications Workshop (AIHA 2020).

Understanding Event Boundaries for Egocentric Activity Recognition from Photo-Streams, by authors Alejandro Cartas, Estefania Talavera, Petia Radeva and Mariella Dimiccoli (Uniersitat de Barcelona & CVC), presented at the Workshop on Applications of Egocentric Vision (EgoApp).

 

Mohamed Ali Souibgui: ICPR2020 Best Student Paper Award

CVC PhD Student Mohamed Ali Souibgui won the ICPR2020 Best Student Paper Award of the Document and Media Analysis Track for his paper “A Few-shot Learning Approach for Historical Ciphered Manuscript Recognition at the 25th International Conference on Pattern Recognition (ICPR 2020). He presented the paper as an oral on January 15th 2021.

Congratulations!

read more
ConferencesCVC NewsPress Release

New advances in fair face recognition algorithms

FairFace Challenge at ECCV 2020(1)

Face recognition has been routinely utilized by both private and governmental organizations around the world. Automatic face recognition can be used for legitimate and beneficial purposes (e.g. to improve security) but at the same time its power and ubiquity heightens a potential negative impact unfair methods can have for the society (e.g. discrimination against ethnic minorities). Although not sufficient, a necessary condition for a legitimate deployment of face recognition algorithms is equal accuracy for all demographic groups.

With this purpose in mind, researchers from the Human Pose Recovery and Behavior Analysis Group at the Computer Vision Center (CVC) – University of Barcelona, led by Dr. Sergio Escalera, organized a challenge within the European Conference of Computer Vision (ECCV) 2020. The results, recently published in Computer Vision – ECCV 2020 Workshops, evaluated the accuracy and bias in gender and skin colour of the submitted algorithms by the participants on the task of face verification in the presence of other confounding attributes.

The challenge was a success since “it attracted 151 participants, who made more than 1.8K submissions in total, exceeding our expectations regarding the number of participants and submissions” explained Dr. Sergio Escalera.

The participants used an image dataset not balanced, which simulates a real world scenario where AI-based models supposed to be trained and evaluated on imbalanced data (considerably more white males that dark females). In total, thy worked with 152,917 images from 6,139 identities.

The images were annotated for two protected attributes: gender and skin colour; and five legitimate attributes: age group (0-34, 35-64, 65+), head pose (frontal, other), image source (still image, video frame), wearing glasses and a bounding box size.

The obtained results were very promising. Top winning solutions exceeded 0.999 of accuracy while achieving very low scores in the proposed bias metrics, which can be considered a step toward the development of fairer face recognition methods. The analysis of top-10 teams shows higher false positive rates for females with dark skin tone and for samples where both individuals wear glasses. In contrast there were higher false negative rates for males with light skin tone and for samples where both individuals are younger than 35 years. “This was not a surprise as the adopted dataset was not balanced with respect to different demographic attributes. However, it shows that overall accuracy is not enough when the goal is to build fair face recognition methods, and that future works on the topic must take into account accuracy and bias mitigation together”, concluded Dr. Julio C. S. Jacques Jr, researcher at the CVC and at the Universitat Oberta de Catalunya.

Reference:

Sixta T., Jacques Junior J.C.S., Buch-Cardona P., Vazquez E., Escalera S. (2020) FairFace Challenge at ECCV 2020: Analyzing Bias in Face Recognition. Computer Vision – ECCV 2020 Workshops. ECCV 2020. Lecture Notes in Computer Science, vol 12540. Springer, Cham. DOI: 10.1007/978-3-030-65414-6_32

In the media

Diario Libre: Detectan algoritmos más precisos de reconocimiento facial según tono de piel

El Desconcierto: Inteligencia artificial: Detectan algoritmos más precisos de reconocimiento facial según tono de piel

Ámbito: Detectan algoritmos que diferencian el tono de piel en el reconocimiento facial

Diario San Rafael: Detectan algoritmos que diferencian el tono de piel en el reconocimiento facial

Mirage news: New advances in detection of bias in face recognition algorithms

Marktechpost: Researchers From Computer Vision Center (CVC) And The University Of Barcelona Conducted A Study That Results In Improved Accuracy On Face Verification Tasks In The Presence Of Other Confounding Attributes

read more
ConferencesCVC News

3 CVC papers accepted at this year’s ECCV

eccv

3 CVC papers have been accepted at the 16th European Conference on Computer Vision (ECCV) that, this year, will be held online from the 23 to the 28th of September. ECCV is one of the world’s top academic conferences in the image analysis area. Congratulations to the authors!

The papers can be accessed here:

Lei Kang, Pau Riba, Yaxing Wang, Marçal Rusiñol, Alicia Fornés, Mauricio Villegas (2020): GANwriting: Content-Conditioned Generation of Styled Handwritten Word Images

Raul Gomez, Jaume Gibert, Lluis Gomez, Dimosthenis Karatzas (2020): Location Sensitive Image Retrieval and Tagging

Hugo Bertiche, Meysam Madadi, Sergio Escalera (2019): CLOTH3D: Clothed 3D Humans

read more
ConferencesCVC News

CVC Researchers at CVPR 2020

cvpr_virtual

CVC researchers have presented a total of 7 papers at this year’s Computer Vision and Pattern Recognition conference (CVPR 2020), the world’s biggest gathering on Computer Vision Science. Like so many events impacted by the COVID-19 pandemic, this year’s edition has taken place virtually from June 14 to June 19th.

CVC’s presence at CVPR 2020 has been the following:

Accepted Papers at CVPR (Main Conference):

  1. Wang, A. González-Garcia, D. Berga, L. Herranz, F. Shahbaz Khan, J. van de Weijer (2020): MineGAN: effective knowledge transfer from GANs to target domains with few images
  2. Porzi, M. Hofinger, I. Ruiz, J. Serrat, S. Rota Bulò, P. Kontschieder (2020): Learning Multi-Object Tracking and Segmentation from Automatic Annotations
  3. Oguz Yazici, A. Gonzalez-Garcia, A. Ramisa, B. Twardowski, J. van de Weijer (2020): Orderless Recurrent Models for Multi-label Classification
  4. Yu, B. Twardowski, X. Liu, L. Herranz, K. Wang, Y. Cheng, S. Jui, J. van de Weijer (2020): Semantic Drift Compensation for Class-Incremental Learning
  5. Wang, S. Khan, A. Gonzalez-Garcia, J. van de Weijer, F. Shahbaz Khan (2020): Semi-supervised Learning for Few-shot Image-to-Image Translation
  6. Sudhakaran, S. Escalera, O. Lanz (2020): Gate-Shift Networks for Video Action Recognition

Invited Talks at Workshops:

Invited Talk by Dr. Dimosthenis Karatzas on “Scene Text VQA: Modelling the interplay between visual and textual information” at the Visual Question Answering and Dialog Workshop, CVPR 2020. Watch the talk here

Invited Talk by Dr. Josep Lladós on “Reading of Population Records: A Digital Twin of the Past Societies” at the Text and Documents in the Deep Learning Era Workshop, CVPR 2020.

Tutorials:

Edgar Riba, Vassileios Balntas, Dmytro Mishkin: Local Features: From SIFT to Differentiable Methods. CVPR 2020 Tutorial.

read more
ConferencesCVC News

A total of 7 CVPRs 2020 accepted at CVC

CVPR_pic

There’s no doubt that this year’s events will be quite different to past years, however, we are very pleased to announce that a total of 7 papers have been accepted from CVC people to this year’s CVPR. Congratulations to the authors!

The papers can be accessed here (list isn’t ready with all the papers, will be updated as soon as they are published in arxiv):

Y. Wang, A. González-Garcia, D. Berga, L. Herranz, F. Shahbaz Khan, J. van de Weijer (2020): MineGAN: effective knowledge transfer from GANs to target domains with few images

L. Porzi, M. Hofinger, I. Ruiz, J. Serrat, S. Rota Bulò, P. Kontschieder (2020): Learning Multi-Object Tracking and Segmentation from Automatic Annotations

V. Oguz Yazici, A. Gonzalez-Garcia, A. Ramisa, B. Twardowski, J. van de Weijer (2020): Orderless Recurrent Models for Multi-label Classification

L. Yu, B. Twardowski, X. Liu, L. Herranz, K. Wang, Y. Cheng, S. Jui, J. van de Weijer (2020): Semantic Drift Compensation for Class-Incremental Learning

Y. Wang, S. Khan, A. Gonzalez-Garcia, J. van de Weijer, F. Shahbaz Khan (2020): Semi-supervised Learning for Few-shot Image-to-Image Translation

S. Sudhakaran, S. Escalera, O. Lanz (2020): Gate-Shift Networks for Video Action Recognition

 

read more
1 2 3 5
Page 1 of 5