[CFP] SHL Challenge 2026 (Locomotion activity recognition challenge)

[pa.l...@gmail.com]

Dear Colleagues,



The Wearable Technologies Lab at the University of Sussex is glad to inform you about the fourth SHL Activity Recognition Challenge. 

                                           http://www.shl-dataset.org/challenge-2026/

This eighth edition of the challenge follows on our very successful 2018, 2019, 2020, 2021, 2023, 2024 and 2025 challenges, which together saw the participation of more than 140 teams and 300 researchers.

Motivated by the growing interest in foundation models, particularly large language models, this year’s edition focuses on their application to transportation mode recognition. The goal is to accurately recognize eight modes of locomotion and transportation in a user-independent manner. In particular, the challenge aims to explore novel ways of adapting, combining, and interfacing foundation models with sensor data to improve recognition performance.
Participants are required to design a complete pipeline that leverages existing pre-trained foundation models to process sensor data and predict activity labels. In the 2025 Challenge, some foundation models have been initially explored for transportation mode recognition, including time-series models such as MOMENT and Chronos, vision models such as Flamingo, and language models such as BERT, used in either pre-trained or retrained settings. This year’s challenge particularly encourages the exploration of new or underutilized foundation models, especially those not investigated in previous editions. Foundation models must be used in a frozen manner, i.e., without any fine-tuning or retraining. However, participants may train lightweight, task-specific components (e.g., classification heads) on top of frozen foundation models, without updating the model parameters.

The provided dataset includes training, validation, and testing sets. Each submission should implement an end-to-end algorithmic pipeline that utilizes foundation models to process sensor data, build predictive models, and output the recognized activities. Participants are also required to submit a technical paper describing their methodology and development process. Accepted contributions will be presented at a special session of the HASCA Workshop at Ubicomp 2026 and included in the adjunct proceedings.

Data to be released on 10.05.2026.

**** Deadlines  *****

• Registration via email: as soon as possible, but not later than 30.05.2026
• Challenge duration: 10.05.2026 – 30.06.2026
• Submission deadline: 30.06.2026
• HASCA-SHL paper submission: 04.07.2026
• HASCA-SHL review notification: 15.07.2026
• HASCA-SHL camera ready submission: 22.07.2026
• UbiComp early-bird registration: TBD
• HASCA workshop: 11-12 October (TBC), 2026 in Shanghai, China
• Release of the ground-truth of the test data: TBD

**** Registration ****
Each team should send a registration email to shldataset...@gmail.com as soon as possible but not later than 15.05.2026, stating the:
• The name of the team
• The names of the participants in the team
• The organization/company (individuals are also encouraged)
• The contact person with email address

****  HASCA Workshop Submission ****

To be part of the final ranking, participants will be required to submit a detailed paper to the HASCA workshop. The paper should contain technical description of the processing pipeline, the algorithms and the results achieved during the development/training phase. The submissions must follow the HASCA format, but with a page limit between 3 and 6 pages.

**** Prizes ****
1.    800 £
2.    400 £
3.    200 £

**** More details ****
http://www.shl-dataset.org/challenge-2026/


**** Contact: ****


All inquiries should be directed to: shldataset...@gmail.com