BMVA Symposium on Human Behaviour Understanding: From Models to Real-World Deployment.

One Day Meeting: BMVA Symposium on Human Behaviour Understanding: From Models to Real-World Deployment.

Wednesday 30 September 2026

Chairs: Jingjing Liu (University of Bristol), Qianhui Men (University of Bristol), Guanxiong Sun (Queen’s University Belfast), Xudong Ma (University of Oxford), Hongbo Bo (University of Bristol), Dr Amirhossein Dadashzadeh (University of Bristol)

Please register to attend on this link:    Register Here to Attend
Sign up for an Expression of Interest to Present via this link: (Deadline 15th July)    Register Here to present

Queries? contact the Meeting's Organiser Andrew Gilbert here

Invited Speakers

Overview

Recent progress in computer vision, multimodal AI, and large-scale learning has created new opportunities for understanding human behaviour from visual and multimodal data. At the same time, significant challenges remain in robustness, generalisation, interpretability, privacy, reliability, and deployment in unconstrained real-world settings. This symposium aims to bring together researchers working on both methodological advances and real-world applications, spanning computer vision, multimodal learning, embodied AI, healthcare, robotics, and assistive technologies.

Schedule

Start   End   Title
09:30   09:50   Registration & Coffee
09:50   10:30   Invited Keynote Speaker – Hyung Jin Chang (University of Birmingham)
10:30   11:00   Accepted Talks Pt. 1 (2 × 15 min: 12 min presentation + 3 min Q&A)
11:00   11:30   Coffee Break
11:30   12:00   Accepted Talks Pt. 2 (2 × 15 min: 12 min presentation + 3 min Q&A)
12:00   12:40   Invited Keynote Speaker – Hubert P. H. Shum (Durham University)
12:40   13:40   Lunch
13:40   14:20   Invited Keynote Speaker – Steven McDonagh (University of Edinburgh)
14:20   15:05   Accepted Talks Pt. 3 (3 × 15 min: 12 min presentation + 3 min Q&A)
15:05   15.35   Coffee Break
15:35   16:15   Invited Keynote Speaker – Dandan Zhang (Imperial College London)
16:15   16:55   Invited Keynote Speaker – Tengda Han (Google DeepMind / University of Oxford)
16:55   17:00   Closing Remarks

Accepted Talks

12 min each + 3 mins questions

Accepted Talks Part 1:

Philip Birch – Towards Robust Human Behaviour Understanding in Unconstrained Environments

Hyun Jin Cho – Amputated Joint Aware 3D Human Mesh Recovery: Toward Inclusive Body Understanding

Accepted Talks Part 2:

Qiushuo Cheng – Towards Vision-Based Quantification of Parkinsonian Turning Motion

Lewis Dickson – More Than a Hand-Waving Argument: Infant Hand Motion from Archival Video with SAM 3 and Optical Flow

Accepted Talks Part 3:

Alzbeta Manova – Goals, Obstacles, and Intentions: Hand Motion Perceptual Failures in Vision-Language Models

Mahrukh Awan – Language-Grounded Detection of Fine-Grained Nonverbal Interactions

Xu Dong – Moving Beyond More Views: Redundancy-Aware Ego–Exo Fusion for Proficiency Estimation

Poster Presentations:

Antonius Bima Murti Wijaya – CrowdTraj: A Benchmark for Automatic Trajectory Predictions Dataset in Crowd Scenes

Sadegh Rahmani – From Human Perception to Human-Inspired Egocentric Action Understanding

Damith Senadeera – From Video-Only to Audio-Visual Violence Detection using Efficient and Adaptive State-Space Models

Oliver Hixon-Fisher – Pose Reconstruction from Sparse Input for VR Avatars

Fatemeh Rahimi – A Multimodal Theory-of-Mind Framework for Trust-Aware HRI

Jordan Williams – Ourobor.Os – Adaptive Human Behaviour, Neurodivergence, Creativity & Cross-Sector Deployment

Important: This is an in-person event, with no virtual attendance option. We kindly ask all presenters to join us at the British Computer Society on the day.

Meeting Location

The meeting will take place at:

British Computer Society (BCS), 25 Copthall Avenue, London EC2R 7BP

Registration

We keep the cost of attending these events as low as possible to ensure no barriers from the whole computer vision community attending. The registration costs are as follows

Please register via charitysuite on this link:    Register Here