Alert button
Picture for Albert Clapés

Albert Clapés

Alert button

SoccerNet 2023 Challenges Results

Sep 12, 2023
Anthony Cioppa, Silvio Giancola, Vladimir Somers, Floriane Magera, Xin Zhou, Hassan Mkhallati, Adrien Deliège, Jan Held, Carlos Hinojosa, Amir M. Mansourian, Pierre Miralles, Olivier Barnich, Christophe De Vleeschouwer, Alexandre Alahi, Bernard Ghanem, Marc Van Droogenbroeck, Abdullah Kamal, Adrien Maglo, Albert Clapés, Amr Abdelaziz, Artur Xarles, Astrid Orcesi, Atom Scott, Bin Liu, Byoungkwon Lim, Chen Chen, Fabian Deuser, Feng Yan, Fufu Yu, Gal Shitrit, Guanshuo Wang, Gyusik Choi, Hankyul Kim, Hao Guo, Hasby Fahrudin, Hidenari Koguchi, Håkan Ardö, Ibrahim Salah, Ido Yerushalmy, Iftikar Muhammad, Ikuma Uchida, Ishay Be'ery, Jaonary Rabarisoa, Jeongae Lee, Jiajun Fu, Jianqin Yin, Jinghang Xu, Jongho Nang, Julien Denize, Junjie Li, Junpei Zhang, Juntae Kim, Kamil Synowiec, Kenji Kobayashi, Kexin Zhang, Konrad Habel, Kota Nakajima, Licheng Jiao, Lin Ma, Lizhi Wang, Luping Wang, Menglong Li, Mengying Zhou, Mohamed Nasr, Mohamed Abdelwahed, Mykola Liashuha, Nikolay Falaleev, Norbert Oswald, Qiong Jia, Quoc-Cuong Pham, Ran Song, Romain Hérault, Rui Peng, Ruilong Chen, Ruixuan Liu, Ruslan Baikulov, Ryuto Fukushima, Sergio Escalera, Seungcheon Lee, Shimin Chen, Shouhong Ding, Taiga Someya, Thomas B. Moeslund, Tianjiao Li, Wei Shen, Wei Zhang, Wei Li, Wei Dai, Weixin Luo, Wending Zhao, Wenjie Zhang, Xinquan Yang, Yanbiao Ma, Yeeun Joo, Yingsen Zeng, Yiyang Gan, Yongqiang Zhu, Yujie Zhong, Zheng Ruan, Zhiheng Li, Zhijian Huang, Ziyu Meng

Figure 1 for SoccerNet 2023 Challenges Results
Figure 2 for SoccerNet 2023 Challenges Results
Figure 3 for SoccerNet 2023 Challenges Results
Figure 4 for SoccerNet 2023 Challenges Results

The SoccerNet 2023 challenges were the third annual video understanding challenges organized by the SoccerNet team. For this third edition, the challenges were composed of seven vision-based tasks split into three main themes. The first theme, broadcast video understanding, is composed of three high-level tasks related to describing events occurring in the video broadcasts: (1) action spotting, focusing on retrieving all timestamps related to global actions in soccer, (2) ball action spotting, focusing on retrieving all timestamps related to the soccer ball change of state, and (3) dense video captioning, focusing on describing the broadcast with natural language and anchored timestamps. The second theme, field understanding, relates to the single task of (4) camera calibration, focusing on retrieving the intrinsic and extrinsic camera parameters from images. The third and last theme, player understanding, is composed of three low-level tasks related to extracting information about the players: (5) re-identification, focusing on retrieving the same players across multiple views, (6) multiple object tracking, focusing on tracking players and the ball through unedited video streams, and (7) jersey number recognition, focusing on recognizing the jersey number of players from tracklets. Compared to the previous editions of the SoccerNet challenges, tasks (2-3-7) are novel, including new annotations and data, task (4) was enhanced with more data and annotations, and task (6) now focuses on end-to-end approaches. More information on the tasks, challenges, and leaderboards are available on https://www.soccer-net.org. Baselines and development kits can be found on https://github.com/SoccerNet.

Viaarxiv icon

Gloss-free Sign Language Translation: Improving from Visual-Language Pretraining

Jul 27, 2023
Benjia Zhou, Zhigang Chen, Albert Clapés, Jun Wan, Yanyan Liang, Sergio Escalera, Zhen Lei, Du Zhang

Figure 1 for Gloss-free Sign Language Translation: Improving from Visual-Language Pretraining
Figure 2 for Gloss-free Sign Language Translation: Improving from Visual-Language Pretraining
Figure 3 for Gloss-free Sign Language Translation: Improving from Visual-Language Pretraining
Figure 4 for Gloss-free Sign Language Translation: Improving from Visual-Language Pretraining

Sign Language Translation (SLT) is a challenging task due to its cross-domain nature, involving the translation of visual-gestural language to text. Many previous methods employ an intermediate representation, i.e., gloss sequences, to facilitate SLT, thus transforming it into a two-stage task of sign language recognition (SLR) followed by sign language translation (SLT). However, the scarcity of gloss-annotated sign language data, combined with the information bottleneck in the mid-level gloss representation, has hindered the further development of the SLT task. To address this challenge, we propose a novel Gloss-Free SLT based on Visual-Language Pretraining (GFSLT-VLP), which improves SLT by inheriting language-oriented prior knowledge from pre-trained models, without any gloss annotation assistance. Our approach involves two stages: (i) integrating Contrastive Language-Image Pre-training (CLIP) with masked self-supervised learning to create pre-tasks that bridge the semantic gap between visual and textual representations and restore masked sentences, and (ii) constructing an end-to-end architecture with an encoder-decoder-like structure that inherits the parameters of the pre-trained Visual Encoder and Text Decoder from the first stage. The seamless combination of these novel designs forms a robust sign language representation and significantly improves gloss-free sign language translation. In particular, we have achieved unprecedented improvements in terms of BLEU-4 score on the PHOENIX14T dataset (>+5) and the CSL-Daily dataset (>+3) compared to state-of-the-art gloss-free SLT methods. Furthermore, our approach also achieves competitive results on the PHOENIX14T dataset when compared with most of the gloss-based methods. Our code is available at https://github.com/zhoubenjia/GFSLT-VLP.

* Accepted to ICCV'23 
Viaarxiv icon

SoccerNet 2022 Challenges Results

Oct 05, 2022
Silvio Giancola, Anthony Cioppa, Adrien Deliège, Floriane Magera, Vladimir Somers, Le Kang, Xin Zhou, Olivier Barnich, Christophe De Vleeschouwer, Alexandre Alahi, Bernard Ghanem, Marc Van Droogenbroeck, Abdulrahman Darwish, Adrien Maglo, Albert Clapés, Andreas Luyts, Andrei Boiarov, Artur Xarles, Astrid Orcesi, Avijit Shah, Baoyu Fan, Bharath Comandur, Chen Chen, Chen Zhang, Chen Zhao, Chengzhi Lin, Cheuk-Yiu Chan, Chun Chuen Hui, Dengjie Li, Fan Yang, Fan Liang, Fang Da, Feng Yan, Fufu Yu, Guanshuo Wang, H. Anthony Chan, He Zhu, Hongwei Kan, Jiaming Chu, Jianming Hu, Jianyang Gu, Jin Chen, João V. B. Soares, Jonas Theiner, Jorge De Corte, José Henrique Brito, Jun Zhang, Junjie Li, Junwei Liang, Leqi Shen, Lin Ma, Lingchi Chen, Miguel Santos Marques, Mike Azatov, Nikita Kasatkin, Ning Wang, Qiong Jia, Quoc Cuong Pham, Ralph Ewerth, Ran Song, Rengang Li, Rikke Gade, Ruben Debien, Runze Zhang, Sangrok Lee, Sergio Escalera, Shan Jiang, Shigeyuki Odashima, Shimin Chen, Shoichi Masui, Shouhong Ding, Sin-wai Chan, Siyu Chen, Tallal El-Shabrawy, Tao He, Thomas B. Moeslund, Wan-Chi Siu, Wei Zhang, Wei Li, Xiangwei Wang, Xiao Tan, Xiaochuan Li, Xiaolin Wei, Xiaoqing Ye, Xing Liu, Xinying Wang, Yandong Guo, Yaqian Zhao, Yi Yu, Yingying Li, Yue He, Yujie Zhong, Zhenhua Guo, Zhiheng Li

Figure 1 for SoccerNet 2022 Challenges Results
Figure 2 for SoccerNet 2022 Challenges Results
Figure 3 for SoccerNet 2022 Challenges Results
Figure 4 for SoccerNet 2022 Challenges Results

The SoccerNet 2022 challenges were the second annual video understanding challenges organized by the SoccerNet team. In 2022, the challenges were composed of 6 vision-based tasks: (1) action spotting, focusing on retrieving action timestamps in long untrimmed videos, (2) replay grounding, focusing on retrieving the live moment of an action shown in a replay, (3) pitch localization, focusing on detecting line and goal part elements, (4) camera calibration, dedicated to retrieving the intrinsic and extrinsic camera parameters, (5) player re-identification, focusing on retrieving the same players across multiple views, and (6) multiple object tracking, focusing on tracking players and the ball through unedited video streams. Compared to last year's challenges, tasks (1-2) had their evaluation metrics redefined to consider tighter temporal accuracies, and tasks (3-6) were novel, including their underlying data and annotations. More information on the tasks, challenges and leaderboards are available on https://www.soccer-net.org. Baselines and development kits are available on https://github.com/SoccerNet.

* Accepted at ACM MMSports 2022 
Viaarxiv icon

Video Transformers: A Survey

Jan 16, 2022
Javier Selva, Anders S. Johansen, Sergio Escalera, Kamal Nasrollahi, Thomas B. Moeslund, Albert Clapés

Figure 1 for Video Transformers: A Survey
Figure 2 for Video Transformers: A Survey
Figure 3 for Video Transformers: A Survey
Figure 4 for Video Transformers: A Survey

Transformer models have shown great success modeling long-range interactions. Nevertheless, they scale quadratically with input length and lack inductive biases. These limitations can be further exacerbated when dealing with the high dimensionality of video. Proper modeling of video, which can span from seconds to hours, requires handling long-range interactions. This makes Transformers a promising tool for solving video related tasks, but some adaptations are required. While there are previous works that study the advances of Transformers for vision tasks, there is none that focus on in-depth analysis of video-specific designs. In this survey we analyse and summarize the main contributions and trends for adapting Transformers to model video data. Specifically, we delve into how videos are embedded and tokenized, finding a very widspread use of large CNN backbones to reduce dimensionality and a predominance of patches and frames as tokens. Furthermore, we study how the Transformer layer has been tweaked to handle longer sequences, generally by reducing the number of tokens in single attention operation. Also, we analyse the self-supervised losses used to train Video Transformers, which to date are mostly constrained to contrastive approaches. Finally, we explore how other modalities are integrated with video and conduct a performance comparison on the most common benchmark for Video Transformers (i.e., action classification), finding them to outperform 3D CNN counterparts with equivalent FLOPs and no significant parameter increase.

Viaarxiv icon

Dyadformer: A Multi-modal Transformer for Long-Range Modeling of Dyadic Interactions

Sep 20, 2021
David Curto, Albert Clapés, Javier Selva, Sorina Smeureanu, Julio C. S. Jacques Junior, David Gallardo-Pujol, Georgina Guilera, David Leiva, Thomas B. Moeslund, Sergio Escalera, Cristina Palmero

Figure 1 for Dyadformer: A Multi-modal Transformer for Long-Range Modeling of Dyadic Interactions
Figure 2 for Dyadformer: A Multi-modal Transformer for Long-Range Modeling of Dyadic Interactions
Figure 3 for Dyadformer: A Multi-modal Transformer for Long-Range Modeling of Dyadic Interactions
Figure 4 for Dyadformer: A Multi-modal Transformer for Long-Range Modeling of Dyadic Interactions

Personality computing has become an emerging topic in computer vision, due to the wide range of applications it can be used for. However, most works on the topic have focused on analyzing the individual, even when applied to interaction scenarios, and for short periods of time. To address these limitations, we present the Dyadformer, a novel multi-modal multi-subject Transformer architecture to model individual and interpersonal features in dyadic interactions using variable time windows, thus allowing the capture of long-term interdependencies. Our proposed cross-subject layer allows the network to explicitly model interactions among subjects through attentional operations. This proof-of-concept approach shows how multi-modality and joint modeling of both interactants for longer periods of time helps to predict individual attributes. With Dyadformer, we improve state-of-the-art self-reported personality inference results on individual subjects on the UDIVA v0.5 dataset.

* Accepted to the 2021 ICCV Workshop on Understanding Social Behavior in Dyadic and Small Group Interactions 
Viaarxiv icon

Deep learning with self-supervision and uncertainty regularization to count fish in underwater images

Apr 30, 2021
Penny Tarling, Mauricio Cantor, Albert Clapés, Sergio Escalera

Figure 1 for Deep learning with self-supervision and uncertainty regularization to count fish in underwater images
Figure 2 for Deep learning with self-supervision and uncertainty regularization to count fish in underwater images
Figure 3 for Deep learning with self-supervision and uncertainty regularization to count fish in underwater images
Figure 4 for Deep learning with self-supervision and uncertainty regularization to count fish in underwater images

Effective conservation actions require effective population monitoring. However, accurately counting animals in the wild to inform conservation decision-making is difficult. Monitoring populations through image sampling has made data collection cheaper, wide-reaching and less intrusive but created a need to process and analyse this data efficiently. Counting animals from such data is challenging, particularly when densely packed in noisy images. Attempting this manually is slow and expensive, while traditional computer vision methods are limited in their generalisability. Deep learning is the state-of-the-art method for many computer vision tasks, but it has yet to be properly explored to count animals. To this end, we employ deep learning, with a density-based regression approach, to count fish in low-resolution sonar images. We introduce a large dataset of sonar videos, deployed to record wild mullet schools (Mugil liza), with a subset of 500 labelled images. We utilise abundant unlabelled data in a self-supervised task to improve the supervised counting task. For the first time in this context, by introducing uncertainty quantification, we improve model training and provide an accompanying measure of prediction uncertainty for more informed biological decision-making. Finally, we demonstrate the generalisability of our proposed counting framework through testing it on a recent benchmark dataset of high-resolution annotated underwater images from varying habitats (DeepFish). From experiments on both contrasting datasets, we demonstrate our network outperforms the few other deep learning models implemented for solving this task. By providing an open-source framework along with training data, our study puts forth an efficient deep learning template for crowd counting aquatic animals thereby contributing effective methods to assess natural populations from the ever-increasing visual data.

* 22 pages, 6 figures, submitted to indexed journal 
Viaarxiv icon

Context-Aware Personality Inference in Dyadic Scenarios: Introducing the UDIVA Dataset

Dec 28, 2020
Cristina Palmero, Javier Selva, Sorina Smeureanu, Julio C. S. Jacques Junior, Albert Clapés, Alexa Moseguí, Zejian Zhang, David Gallardo, Georgina Guilera, David Leiva, Sergio Escalera

Figure 1 for Context-Aware Personality Inference in Dyadic Scenarios: Introducing the UDIVA Dataset
Figure 2 for Context-Aware Personality Inference in Dyadic Scenarios: Introducing the UDIVA Dataset
Figure 3 for Context-Aware Personality Inference in Dyadic Scenarios: Introducing the UDIVA Dataset
Figure 4 for Context-Aware Personality Inference in Dyadic Scenarios: Introducing the UDIVA Dataset

This paper introduces UDIVA, a new non-acted dataset of face-to-face dyadic interactions, where interlocutors perform competitive and collaborative tasks with different behavior elicitation and cognitive workload. The dataset consists of 90.5 hours of dyadic interactions among 147 participants distributed in 188 sessions, recorded using multiple audiovisual and physiological sensors. Currently, it includes sociodemographic, self- and peer-reported personality, internal state, and relationship profiling from participants. As an initial analysis on UDIVA, we propose a transformer-based method for self-reported personality inference in dyadic scenarios, which uses audiovisual data and different sources of context from both interlocutors to regress a target person's personality traits. Preliminary results from an incremental study show consistent improvements when using all available context information.

* Accepted to the 11th International Workshop on Human Behavior Understanding workshop at Winter Conference on Applications of Computer Vision 2021 
Viaarxiv icon