Deep Driving: Learning Representations for Intelligent Vehicles
Together with University of California at Berkeley, University of Jenna, NICTA and Daimler we are organising a workshop in representation learning at the IEEE Intelligent Vehicles Symposium in Gothenburg, Sweden. Vision is a rich and…
Monitoring Vehicle Occupants
Visual Monitoring of Driver and Passenger Control Panel Interactions Researchers Toby Perrett and Prof. Majid Mirmehdi Overview Advances in vehicular technology have resulted in more controls being incorporated in cabin designs. We present a system…
High Frame Rate Video
As the demand for higher quality and more immersive video content increases, the need to extend the current video parameter space of spatial resolutions and display sizes, to include, among other things, a wider colour gamut, higher dynamic range…
Fake eyes? – How Eyespots work
BVI Seminar with speaker Dr. Karin Kjernsmo from University of Bristol.
What’s on TV: A Large-Scale Quantitative Characterisation of Modern Broadcast Video Content
What does one year of modern broadcast video look like? This project analyses the spatial and temporal diversity in contemporary video to make inferences about testing new video technologies.
BVI on The One Show
In an episode of The One Show which aired on the 15th April, BVI contributed to an item demonstrating the dangers of texting while driving. Recent studies have indicated that reading and even writing text messages…
Cuttlefish vision in a 3-D world
BVI Seminar with speaker Professor Daniel Osorio from University of Sussex.
You-Do, I-Learn: Egocentric Unsupervised Discovery of Objects and their Modes of Interaction
Arxiv version of our paper 2016 CVIU paper is now available at: http://arxiv.org/abs/1510.04862v2
Adaptive Resolution Intra Coding
Adaptive resolution coding of intra frames for improved rate distortion performance at low bitrates.
Remote Pulmonary Function Testing using a Depth Sensor
We propose a remote non-invasive approach to Pulmonary Function Testing using a time-of-flight depth sensor (Microsoft Kinect V2), and correlate our results to clinical standard spirometry results. Given point clouds, we approximate 3D models of the subject’s chest, estimate…