One Day Meeting: BMVA Symposium: Robotics Foundation & World Models

Wednesday 30 October 2024

Chairs: Rudra Poudel (Toshiba Europe Ltd), Dimitrios Kanoulas (University College London), Maria Koskinopoulou (Heriot-Watt University), Thomas George Thuruthel (University College London), Edward Johns (Imperial College London)

Queries? contact the Meeting's Organiser Andrew Gilbert here

Videos of Talks

On our BMVA YouTube channel there are recorded talks of the slides and speaker from the day here

Programme

Lightning Talks I (12 mins each)

• Xiaofeng Mao et al. DexSkills: Skill Segmentation Using Haptic Data for Learning Autonomous Long-Horizon Robotic Manipulation Tasks
• Vitalis Vosylius et al. Instant Policy: In-Context Imitation Learning via Graph Diffusion

Lightning Talks II (12 mins each)

• Michael Mangan. Opteran: reverse engineering insect brains to provide the software mind that delivers intelligent machines
• Ivan Kapelyukh et al. Foundation Models as Imagination Engines for Robotics
• Xintong Yang et al. GAM: General Affordance-based Manipulation for Contact-rich Object Disentangling Tasks

Posters

• Naoki Kiyohara et al. Learning SDE Solutions with Neural Stochastic Flows
• Nikolaos Tsagkas et al. Click to Grasp: Zero-Shot Precise Manipulation via Visual Diffusion Descriptors
• Kaiwen Cai et al. Self-Adapting Large Visual-Language Models to Edge Devices across Visual Modalities
• Andreas Sochopoulos et al. Learning Deep Dynamical Systems using Stable Neural ODEs
• Miles Hansard. Polycube objects for robotic grasping research
• Edgar Sucar et al. Simple4D: A Simple Model 4 Dynamic Scene Reconstruction
• B. K. Dhillon1 et al. Enhancing Accessibility for the Visually Impaired: A Review of Machine Learning Approaches for Vacant Seat Identification in Public Spaces
• Imene Tarakli et al. Sleep-Inspired Consolidation with Human Feedback Learning
• Georgios Papagiannis et al. R+X: Retrieval and Execution from Everyday Human Videos
• Jianhao Jiao et al. LiteVLoc: Map-Lite Visual Localization for Image Goal Navigation
• Markus Buchholz et al. Motion Control for Tethered Multi-Robot Systems in Marine Environments with Disturbances

To contact the chairs

Rudra P.K. Poudel, Toshiba Europe Ltd, rudra.poudel@gmail.com

Dimitrios Kanoulas, University College London, d.kanoulas@ucl.ac.uk

Maria Koskinopoulou, Heriot-Watt University, M.Koskinopoulou@hw.ac.uk

Thomas George Thuruthel, University College London, t.thuruthel@ucl.ac.uk

Edward Johns, Imperial College London, e.johns@imperial.ac.uk,

Overview of Technical Meeting

This meeting will bring together academic and industry presentations, researchers interested in all aspects of learning world models and foundational models for robotics, including but not limited to,

• World models,
• Robotic foundation models,
• Efficient foundation models,
• Planning,
• Manipulation,
• Navigation,
• Multimodal representation learning,
• Vision-and-language for robotics,
• etc.

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

• All Attendees: £30 Including lunch and refreshments for the day

Queries? contact the Meeting's Organiser Andrew Gilbert here