Media & podcast
Computer Vision News - August 2020
*NEW* SARAS Challenge features on Best of MIDL 2020
The SARAS-ESAD Challenge we organised for MIDL 2020 (the Medical Imaging with Deep Learning conference), held virtually in Montreal in July 2020, was selected to feature in 'Best of MIDL 2020'.
*NEW* Video recording of the SARAS-ESAD Challenge at MIDL 2020
Complete video recording of the SARAS-ESAD Challenge we organised at MIDL 2020 (the Medical Imaging with Deep Learning conference).
*NEW* Generalized Zero-Shot Learning using Generated Proxy Unseen Samples and Entropy Separation
Supplementary video submitted to MM 2020
*NEW* Official SARAS video
Saras Multirobots-Surgery is the first step toward the next-generation of surgical robotic systems merging artificial intelligence, computer vision, machine learning and cognitive control for smarter and safer hospitals.
The MULTIROBOTS SURGERY platform is the first level of increasing complexity of the robotic system developed in SARAS Project. The Multirobots consist of two parts: a commercial robotic system used by the main surgeon and the teleoperated SARAS assistive robotic arms used by the assistant.
SARAS: Multi-robot Cooperation and Task Planning - Task identification and training data
The actions performed by the assistant surgeon during the MULTIROBOTS-SURGERY will be mathematically described and parameterized in order to be executed and adapted at run time by the SARAS assistive robotic arms.
Work in progress: testing phantoms and SARAS arms - Part 2. This is the expected result of the functioning of the SARAS arms with the new phantoms after one year project.
SARAS MULTIROBOTS platform set-up
Watch the video of our test of the prostate's phantom and of the positioning of the SARAS arms with respect to the phantom and the da Vinci® arms.
Robotics arms for laparoscopic surgery
Venturefest Oxford, September 2019
Incremental Tube Construction for Human Action Detection
Harkirat Behl et al
British Machine Vision Conference 2018
September 2018
Rowing for Parkinson's
De Montfort University technical instructor, Robin Buttery, rowed 3,600 miles across the Indian Ocean as part of a study by Dr Shelly Coe of Oxford Brookes University
BBC Inside Out East Midlands
14 January 2019
The crucial role of theory of mind capabilities
Venturefest Oxford
September 2018
Towards machines that can read your mind
Professorial lecture
Open Lecture Series
Jan 24 2018
Towards machines that can read your mind
Professorial lecture
Facebook Live
Jan 24 2018
Online Real-time Multiple Spatiotemporal Action Localisation and Prediction (ICCV 2017)
Advances in Artificial Intelligence: Gesture and Action Recognition
Risk Roundup LLC
2016 Harvard Statistics invited seminar
Belief functions: past, present and future
July 14, 2016
Deep Learning for Detecting Multiple Space-Time Action Tubes in Videos (BMVC 2016)
UAI 2015 Tutorial
Belief functions for the working scientist
Amsterdam, July 2015
MSc in Computer Vision (2014)
Deformable Part-Based Models for Action Localisation (IJCV 2014)
Introducing Artie - Our humanoid Robothespian robot
Multiple Instance Learning for Action Localisation (BMVC 2012)
Spectral matching of video meshes (INRIA, CVPR 2008)