2013 Intelligent Sensing Summer School

A double interview from the 2013 Intelligent Sensing Summer School: link

Location: Biology Lecture Theatre, G.E. Fogg building

11 June 2013 - DAY 1
9.00-9.30 Opening
Integrative neuromechanics: combining engineering, mathematics, and biology to discover how animals move

Insects on rubber, dogs on springs, and robots in a field: the integrative science of discovering how animals move

Andrew Spence slide
Andrew Spence video
Andrew Spence Andrew Spence Biology, vision & intelligent sensing I
11.00-11.30 Coffee break
Depth, You, and the World

Consumer level depth cameras such as Kinect have changed the landscape of 3D computer vision. In this talk, we will discuss two related approaches that learn to directly infer correspondences between observed depth image pixels and 3D model points, and then use these correspondences to drive an optimization of a generative model to explain the data.

Jamie Shotton slide
Jamie Shotton video
Jamie ShottonJamie Shotton Industry talk (Microsoft Research)
Capturing and modelling 3D data

Passive and active systems for capturing 3D scene-data, and a statistical scene-model discussed in relation to biological and machine vision

Miles Hansard slide
Miles HansardMiles Hansard Biology, vision & intelligent sensing II
14.45-15.00 Presentation of the challenge Challenge
Biology, vision & intelligent sensing
15.00-18.00 Working in groups
12 June 2013 - DAY 2
Analysing language in conversation

Approaches for extracting meaning from conversational language, from deep to shallow and with varying degrees of supervision for interactive dialogue systems, medical therapy assistance and social media analytics

Matthew Purver slide
Matthew Purver video
Matthew Purver Matt Purver Sensing language, audio & music
Making sense of sound and music

Techniques that can separate out different sound sources from a mixture, follow the notes and the beats in a piece of music, or show us the sound in new visual ways to analyze sounds like heartbeats or birdsong, find the music we want in huge collections of music tracks, or to create music in new ways

Mark Plumbley video
Mark Plumbley Mark Plumbley
11.00-11.30 Coffee break
Smart cooperative wireless sensor networks for healthcare and performance monitoring: antennas and radios prospective

Techniques that combine antenna and radio propagation engineering with networking and smart frequency agile communication systems for low-power body-worn or implanted devices coupled with intelligent control algorithm to sense and understand the dynamic human body

Akram Alomainy slide
Akram Alomainy video
Alomainy Akram Akram Alomainy Sensing human body signals
Advanced Cardiovascular Imaging: current state and glimpse into the future

Cardiac image acquisition, display, post-processing and handling of clinical reports

Steffen Petersen video
Steffen Petersen Steffen Petersen Sensing for medical applications
14.30-18.00 Working in groups Challenge
Biology, vision & intelligent sensing
13 June 2013 - DAY 3
8.30 Submission of the challenge results
9.00-9.45 Shining new light on the Earth's surface: hyperscale terrain monitoring and modelling

Recent advances in Geomatics that have transformed the acquisition and modelling of 3D survey data, and facilitated new perspectives on the form and structure of the Earth’s surface and dynamics
James Brasington James Brasington Geography & intelligent sensing
9.45-10.00 Groups 1 & 2 present their results on the challenge
Using insect sensory systems as biological inspiration for complex sensing problems

Insects are masters of multisensory integration: they use polychromatic colour vision, polarisation vision, olfaction and a magnetic compass system to navigate and to explore their surroundings with brains the size of a pinhead. How they do this could provide useful information for computationally simple but robust solutions to complex sensing problems.

Lars Chittka slide
Lars Chittka video
Lars Chittka Lars Chittka Evolutionary ecology & intelligent sensing
10.45-11.00 Groups 3 & 4 present their results on the challenge
11.00-11.30 Coffee break
Probabilistic machine learning models for computer vision and beyond

Exploitation of weakly-supervised, semi-supervised and active learning as well as attributes and zero-shot learning for intelligent sensing applications in computer vision and beyond

Timothy Hospedales slide
Timothy Hospedales Tim Hospedales Machine learning & intelligent sensing
12.15-12.30 Groups 5, 6 & 7 present their results on the challenge
Design for a Darwinian brain

Artificial curiosity and algorithms for making robots creative and imaginative. They make up their own games to play and play them. To do this they co-evolve populations of games and solutions in their brains

Chrisantha Fernando slide
Chrisantha Fernando Chrisantha Fernando Robots & intelligent sensing
Textural features and illumination invariance

Appearance of natural and artificial materials in real scenes is highly dependent on camera position, illumination position and spectrum, which complicate image analysis. These issues are addressed with illumination invariants from a Markovian representation of texture.

Pavel Vacha slide
Pavel Vacha video

Demo: information fusion for localisation systems

Pavel Vácha Pavel Vácha Industry talk
15.30-15.45 Challenge results & award ceremony
15.45-17.00 Third-year PhD students tell their stories and answer to questions from first and second year colleagues
Panel moderators: Evangelos Sariyanidi and Riccardo Mazzon
Panel discussion
Programme organiser Challenge organisers Logistics Sponsors
Hatice Gunes
Tomas Piatrik
Tomas Piatrik
Julie Wall
Julie Wall
Qianni Zhang
Qianni Zhang
Tamer Aboufoul
Akram Alomainy
Richa Bharadwaj
Eliane Bodanese
Simon Emberton
Manik Gupta
Kok Ho Huen
Yue Jia
Nanda Khaorapapong
Evangelos Sariyanidi
Ke Yang
EPSRC logo
CIS logo
Winners of the challenge

1° Prize: Julian Hough, Fabio Poiesi, Evangelos Sariyanidi, Daria Stefic

2° Prize: Sophia Bano, Obaidullah Khalid, Mark Roper, Valerio Targon

3° Prize: Wenzhao Li, Riccardo Mazzon, Xun Xu, Heng Yang

[Flickr album]