Research interests: Computer Vision, Machine Learning, and AI for understanding human behavior (actions/gestures and beyond:
psychological states, intention anticipation, efficient perception via multimodality, pretraining, and multi-task distillation).
Publications (highlighted)
This is a selection of the latest and most relevant publications. For journals, I report Journal Impact Factor (JIF/JCR) as stated in WoS; for peer-reviewed conference proceedings, I report GGS class.
Links point to open-access or publisher pages.
2026
JOURNAL Grosselle R, Lindgaard E, Xarles A, Clapés A, Escalera S, Bak BLV. The feasible-fields method: A deep learning framework for progressive damage modelling of delamination under quasi-static loading. Composites Part B: Engineering, 2027.
JIF 14.0 (JCR 2025, Q1).ScienceDirect
CONF. PROCEEDINGS Xarles A, Escalera S, Clapés A. AdaSpot: Spend Resolution Where It Matters for Precise Event Spotting.
CVPR 2026.
GGS Class 1 (A++). AR 25.42%.
CONF. PROCEEDINGS Pujol-Perich D, Clapés A, Damen D, Escalera S, Wray M. Beyond Caption-Based Queries in Video Moment Retrieval.
CVPR 2026.
GGS Class 1 (A++). AR 25.42%.
2025
CONF. PROCEEDINGS Pujol-Perich D, Escalera S, Clapés A. Sparse-Dense Side-Tuner for efficient Video Temporal Grounding (SDST).
ICCV 2025.
GGS Class 1 (A++). AR 24.03% (source).Paper
CONF. PROCEEDINGS Xarles A, Escalera S, Moeslund TB, Clapés A. Action Valuation in Sports: A Survey. CVPR Workshops 2025.
Workshop of GGS Class 1 (A++).IEEE
CONF. PROCEEDINGS Dalal M, Xarles A, Cioppa A, Giancola S, Van Droogenbroeck M, Ghanem B, Clapés A, Escalera S, Moeslund TB.
Action Anticipation from SoccerNet Football Video Broadcasts. CVPR Workshops 2025.
Workshop of GGS Class 1 (A++).IEEE
CONF. PROCEEDINGS Pujol-Perich D, Clapés A, Escalera S. SADA: Semantic Adversarial Unsupervised Domain Adaptation for Temporal Action Localization.
WACV 2025.
GGS Class 2. AR 37.8% (source).IEEE
2024
CONF. PROCEEDINGS Xarles A, Escalera S, Moeslund TB, Clapés A. T-DEED: Temporal-Discriminability Enhancer Encoder-Decoder for Precise Event Spotting in Sports Videos.
CVPR Workshops 2024.
Workshop of GGS Class 1 (A++).IEEE
CONF. PROCEEDINGS Zhou B, Chen Z, Clapés A, Wan J, Liang Y, Escalera S, Lei Z, Zhang D.
Gloss-free sign language translation: Improving from visual-language pretraining (GFSLT-VLP).
ICCV 2023.
GGS Class 1 (A++). AR 26.15% (source).IEEE
CONF. PROCEEDINGS Xarles A, Escalera S, Moeslund TB, Clapés A. Astra: An action spotting transformer for soccer videos.
MMSports 2023.
ACM
JOURNAL Selva J, Johansen AS, Escalera S, Nasrollahi K, Moeslund TB, Clapés A.
Video Transformers: A Survey. IEEE TPAMI, 2023.
JIF 20.8 (JCR), Q1; position 2/197 (CS AI).IEEE
2022
JOURNAL Tarling P, Cantor M, Clapés A, Escalera S.
Deep learning with self-supervision and uncertainty regularization to count fish in underwater images.
PLoS ONE, 2022.
JIF 3.7 (JCR), Q2.PLOS
2021
CONF. PROCEEDINGS Clapés A, Curto D, Selva J, …, Escalera S. Dyadformer: A Multi-modal Transformer for Long-Range Modeling of Dyadic Interactions.
ICCV Workshops 2021.
GGS Class 1 (A++).IEEE
CONF. PROCEEDINGS Palmero C, Selva J, Smeureanu S, …, Escalera S.
Context-Aware Personality Inference in Dyadic Scenarios: Introducing the UDIVA Dataset.
WACV Workshops 2021.
GGS Class 2.Springer
Earlier years
CONF. PROCEEDINGS Jacques-Junior JCS, Morral C, Escalera S, Clapés A. ChaLearn LAP 2020 Challenge on Identity-preserved Human Detection: Dataset and Results.
FG 2020.
GGS Class 2.IEEE
JOURNAL Ofodile I, Helmi A, Clapés A, …, Anbarjafari G. Action Recognition Using Single-Pixel Time-of-Flight Detection.
Entropy, 2019.
JIF 2.494 (JCR), Q2.MDPI
JOURNAL Clapés A, Pardo À, Pujol-Vila O, Escalera S.
Action detection fusing multiple Kinects and a WIMU: an application to in-home assistive technology for the elderly.
Machine Vision and Applications, 2018.
JIF 1.788 (JCR), Q2.Springer
JOURNAL Palmero C, Clapés A, Bahnsen C, Mogelmose A, Moeslund TB.
Multi-modal RGB-Depth-Thermal Human Body Segmentation.
IJCV, 2016.
JIF 8.222 (JCR), Q1.Springer
CONF. PROCEEDINGS Cepero Á, Clapés A, Escalera S. Automatic non-verbal communication skills analysis: A quantitative evaluation.
AI Communications, 2015.
JIF 0.364 (JCR), Q4.Link
CONF. PROCEEDINGS Reyes M, Clapés A, Ramírez J, Revilla JR, Escalera S.
Automatic digital biometry analysis based on depth maps.
Computers in Industry, 2013.
JIF 1.457 (JCR), Q3.ScienceDirect
JOURNAL Clapés A, Reyes M, Escalera S.
Multi-modal user identification and object recognition surveillance system.
Pattern Recognition Letters, 2013.
JIF 1.062 (JCR), Q3.ScienceDirect
Projects (highlighted)
The following list is a selection of the latest and most relevant projects.
European Projects
Diversibus Viis Plurima Solvo (DVPS) — HORIZON-CL4-2024-HUMAN-03, Horizon Europe,
Grant: 101213369, Period: 01/07/2025–30/06/2029.
Role: project member (UB partnership).
Hacia una IA segura y confiable — Proyectos de Generación de Conocimiento 2022 (Investigación No Orientada), Ministerio de Ciencia e Innovación,
Ref: PID2022-136436NB-I00, Period: 01/09/2023–31/08/2026.
Role: project member.
Regional Projects (Catalonia)
Noves dianes terapèutiques en la depressió: el paper de la percepció situacional — La Marató de TV3 – Salut Mental (2021), Fundació La Marató de TV3,
Ref: 433/U/2022, Period: 15/03/2023–14/03/2026.
Role: project member.
Transference / Industry Collaboration Projects
Multi-Modal Knowledge Distillation for Improved Visual Understanding — Huawei,
Ref: FBG 313528, Period: 05/12/2025–04/12/2027.
Funding: €251,406.Role: Principal Investigator.
A Virtual Clothing Try-On System Based on 3D Body and Garment Reconstruction and Animation — Indústria del Coneixement 2024 – Producte (Modalitat B), Generlitat de Catalunya,
Ref: 2024 PROD 00013, Period: 02/12/2024–01/06/2026.
Funding: €150,000.Role: Project member.
NEFA YouAddict EIT Digital Project — EIT Digital 2024,
Ref: FBG 23395, Period: 01/01/2024–30/06/2024.
Funding: €115,000.Role: Project member.
Research Services on Fashion Recommendation and Retrieval — YouAddict Srl,
Ref: FBG 312945, Period: 06/09/2024–05/03/2025.
Funding: €40,000.Role: Project member.
Consultancy on Milestone AI Research Strategy — Milestone Systems S/A,
Ref: FBG 312249; Period: 08/05/2023–07/12/2023.
Funding: €7,317.Role: Project member.
PhD theses supervision
Javier Selva (University of Barcelona, defended on Dec 19th 2023)
David Pujol-Perich (University of Barcelona, defended on Jun 17th 2026)
Artur Xarles (University of Barcelona, on-going)
Michal Choinski (Industrial PhD (AGAUR): University of Barcelona × Restb.ai, on-going)
Teaching (summary)
University teaching load: 3.375 full-time equivalent years; 135 hours theoretical + 566.5 hours practical
(700+ total hours reported).
Undergraduate teaching (UB): Computer Science / Mathematics degrees (including double degree).
Courses taught include: Programming I, Programming II, Data Structures, Software Design, Integrated Software Project.
Master teaching: Video Analysis at Interuniversity Master in Computer Vision (UAB, UPC, UPF, UB, and UOC) in 2023/24 and 2024/25.
Supervision: 12× BSc theses (TFG) and 4× MSc theses (TFM).
Invited teaching: Université Paris-Saclay (Computer Vision master), two lectures in 2021/22 and 2022/23.
Other merits
Awards
Top reviewer of NeurIPS26 Evaluations & Datasets
Outstanding reviewer of CVPR26 (top 5%, 16K+)
Extraordinary PhD Award — PhD, University of Barcelona (2019).
Dissertation: “Learning to recognize human actions: from hand-crafted to deep-learning based visual representations”.
Cum Laude; International Mention.
Scientific Advisor and Consultant for the BitSpace spin-off in the AI Accelerator programme (Xarxa RDI-IA, Generalitat of Catalonia). Link
Scientific Advisory, AI-Accelerator (BitSpace). Link
(Co-)organization of International Conference Workshops:
2nd International Workshop & Challenge on Subtle Visual Computing @ CVPR26
12th International Workshop on Computer Vision in Sports (CVsports) @ CVPR26
Understanding Social Behavior in Dyadic and Small Group Interactions Challenge @ ICCV21
ChaLearn LAP Workshop @ ICPR16
ChaLearn LAP Workshop @ ECCV16
Organization of Challenges (and associated Workshop):
SoccerNet's Ball Action Anticipation Challenge (associated to CVsports @ CVPR26)
Chalearn's Understanding Social Behavior in Dyadic and Small Group Interactions Challenge (associated to Understanding Social Behavior in Dyadic and Small Group Interactions Challenge @ ICCV21)
Chalearn's Identity-preserved Human Detection
ChaLearn's First Impressions Challenge (associated to ChaLearn LAP Workshop @ ICPR16 and ChaLearn LAP Workshop @ ECCV16)