Technical Programme

Technical Programme

IMVIP 2017 will take place in the IONTAS Lecture Theatre and Foyer from the afternoon of Wednesday 30th August through to the morning of Friday 1st September (inclusive). The welcome reception with be held on the South Campus in Renehan Hall. Lunch will be provided on the Thursday of the conference in the Phoenix University Restaurant.

 

Wednesday 30th August

12noon - 2pm  Registration desk open (Iontas Building Foyer)

2pm - 2.15pm Welcome and Opening Remarks, Prof. Ray O'Neill, Vice-President for Research, Maynooth University.

2.20pm - 3.20pm Keynote - Chair: John McDonald


Thomas Whelan, Oculus Research

Title: Machine Perception

Abstract: The advent of modern virtual reality has brought about a host of new and challenging problems in many fields, most notably in computer vision. The future of this technology is not just restricted to the domain of virtual reality, but has far greater reach into more challenging areas including mixed and augmented reality. The Surreal Vision team at Oculus Research focuses on real-time always on embedded machine perception. This talk will present a brief overview of Oculus Research, some of the machine perception challenges involved in mixed reality and some of the existing published solutions to these problems.


3.20pm - 3.50pm Coffee / Tea Break

3.50pm - 4.50pm Oral Paper Session 1 - Chair: Paul Whelan

  • A Modular Scheme for Artifact Detection in Stereoscopic Omni-Directional Images
    Sebastian Knorr, Simone Croci and Aljosa Smolic
     
  • A Robust Quality Measure for Quality-Guided 2D Phase Unwrapping Algorithms Based on Histogram Processing
    Ambroise Moreau, Matei Mancas and Thierry Dutoit    
     
  • Fast and Accurate Optical Flow based Depth Map Estimation from Light Fields
    Yang Chen, Martin Alain and Aljosa Smolic  

6pm - 8pm Welcome Reception in Renehan Hall with tours of the Russell Library

 

Thursday 31st August

9.20am - 10.40am Oral Paper Session 2 - Chair: Bryan Gardiner

  • Recognising Fine-Grained Actions by Combining Colour, Depth and Flow
    Seán Bruton and Gerard Lacey    
     
  • A Tale of Two Losses: Discriminative Deep Feature Learning for Person Re-Identification
    Alessandro Borgia, Yang Hua and Neil Robertson    
     
  • On using CNN with DCT based Image Data
    Matej Ulicny and Rozenn Dahyot
     
  • Deep convolutional neural networks and digital holographic microscopy for in-focus depth estimation of microscopic objects
    Tomi Pitkaaho, Aki Manninen and Thomas Naughton    

10.40am - 11.40am Poster Pitches / Poster Session 1/ Industry Demos & Coffee - Chair: Kenneth Dawson-Howe

  • Computing the Uncertainty of Motion Fields for Human Activity Analysis
    Jorge S. Marques, António R. Moreira and João M. Lemos    
     
  • Automatic Book Finding on Bookshelves
    Jason Hogan and Kenneth Dawson-Howe    
     
  • Image Fusion of Unregistered Colour Digital Pathology Images
    Wael Saafin, Gerald Schaefer, Miguel Vega, Rafael Molina and Aggelos Katsaggelos    
     
  • High Speed Reconstruction of a Scene Implemented Through Projective Texture Mapping
    William Clifford, Catherine Deegan and Charles Markham    
     
  • Improving The Viola-Jones Face Detection Performance by Using The Brightness Channel in HSV and HLS Colour Spaces
    Inas Altaie, Adrian Clark and Nassr Azeez    
     
  • Tahitian Pearls’ Lustre Assessment
    Gael Mondonneix, Sébastien Chabrier, Jean-Martial Mari, Alban Gabillon and Jean-Pierre Barriot    
     
  • Fast Video Processing Using a Spiral Coordinate System and an Eye Tremor Sampling Scheme
    John Fegan, Sonya Coleman, Dermot Kerr and Bryan Scotney    

11.40am - 12.40pm Oral Paper Session 3 - Chair: Bryan Scotney

  • Automated Identification of Trampoline Skills Using Computer Vision Extracted Pose Estimation
    Paul Connolly, Guenole Silvestre and Chris Bleakley    
     
  • Visual Lecture Summary Using Intensity Correlation Coefficient
    Solomon E. Garber, Luka Milekic, Nick Moran, Aaditya Prakash, Antonella Di Lillo and James A. Storer    
     
  • Automatic Tracking System for Event Detection and Classification in Tennis
    Pushyami Rachapudi, Abhishek Sharma and Navjyoti Singh    

12.40pm - 2pm Lunch @ Phoenix University Restaurant

2pm - 3pm Keynote - Chair: Charles Markham


Maurice Fallon, University of Oxford

Title: Mapping and Tracking for Legged Robots

Abstract: In this talk I will present ongoing research on multi-sensor perception for applications to robotic manipulation, navigation and locomotion. The first topic will focus on creating volumetric maps suitable for collision free motion planning by incorporating humanoid proprioception into a surfel-based dense visual mapping system. Visual SLAM systems are notoriously fragile - either reliant on the presence of structure or assuming no dynamics in the scene. A second topic develops methods for high-frequency legged robot state estimation - combining proprioception, vision and LIDAR. We will describe the various challenges of doing so - including dynamic impacts, poor lighting conditions and various sources of sensor noise. Finally we present initial research on visual manipulator tracking. We explore the problem instead by tracking the robot’s manipulator using dense vision and present initial research showing how we can use discriminative methods to propose candidate solutions. Example demonstrations will be given using some notable platforms including the Boston Dynamics Atlas, NASA Valkyrie and the HyQ quadruped.


3pm - 3.40pm Oral Paper Session 4 - Chair: Barak Pearlmutter

  • Saliency Detection and Object Classification
    Christopher Cooley, Sonya Coleman, Bryan Gardiner and Bryan Scotney    
     
  • IDT Vs L2 Distance for Point Set Registration
    Hana Alghamdi, Mairead Grogan and Rozenn Dahyot

3.40pm - 4.40pm Poster Pitches / Poster Session 2 / Industry Demos & Coffee - Chair: Dermot Kerr

  • Gaussian Random Vector Fields in Trajectory Modelling
    Miguel Barão and Jorge S. Marques    
     
  • Digital holographic sensor network and image analyses for distributed potable water monitoring
    Tomi Pitkaaho, Ville Pitkakangas, Mikko Niemela, Sudheesh K. Rajput, Naveen K. Nishchal and Thomas Naughton    
     
  • Study of imperfect keys to characterise the security of optical encryption
    Lingfei Zhang and Thomas Naughton    
     
  • Towards Dense Collaborative Mapping using RGBD Sensors
    Louis Gallagher and John McDonald    
     
  • Video Based Piano Music Transcription
    Robert McCaffrey and Kenneth Dawson-Howe    
     
  • Structure Based Matching between Aerial and Map Images using Brightness- and Rotation-Invariant Curve Features
    Yoshikatsu Nakajima and Hideo Saito    
     
  • A dataset for Irish Sign Language recognition
    Marlon Oliveira, Houssem Chatbri, Noel E. O'Connor, Alistair Sutherland, Ylva Ferstl, Suzanne Little and Mohamed Farouk    

4.40 - 5.40pm Oral Paper Session 5 - Sonya Coleman

  • Detecting and Tracking Meeting Participants using Motion Heat Maps
    Nahlah Algethami and Sam Redfern    
      
  • Gabor and HOG approach to facial emotion recognition
    Ryan Melaugh, Nazmul Siddique, Sonya Coleman and Pratheepan Yogarajah    
     
  • SimilarityMeasures and the Performance of Biometric Systems
    Inas Altaie, Adrian Clark and Nassr Azeez    
     

6pm - 7pm IPRCS AGM

Thursday Evening - Conference Dinner at Barberstown Castle Hotel, Straffan, Co. Kildare.
Bus departs from Phoenix Restaurant Pickup Point at 7.30pm.

 

Friday 1st September

9.20am - 10.20am Keynote - Chair: Rozenn Dahyot


Stefan Leutenegger, Imperial College London

Title: Spatial Perception for Mobile Robots

Abstract: Mobile robots need dedicated sensing and processing for localisation and mapping as well as scene understanding. Recent years have brought tremendous advances in vision sensors (e.g. RGB-D cameras) and processing power (e.g. GPUs) that have led us to design new algorithms that will empower the next generation of mobile robots. With the arrival of deep learning, we are furthermore now in the position to link respective unprecedented performance in scene understanding with 3D mapping. In this talk, I will go through some recent algorithms and software we have developed as well as their application to mobile robots, including drones.


10.20am - 11.20am Poster Pitches / Poster Session 3 & Coffee - Chair: Charles Markham

  • Extending the Bag-of-Words Representation with Neighboring Local Features and Deep Convolutional Features
    Daniel Manger and Dieter Willersinn    
     
  • Local Shape and Moment Invariant Descriptor for Structured Images
    Elena Ranguelova    
     
  • Open Source Dataset and Deep Learning Models for Online Digit Gesture Recognition on Touchscreens
    Philip Corr, Chris Bleakley and Guenole Silvestre    
     
  • Facial Image Aesthetics Prediction with Visual and Deep CNN Features
    Mohamed Selim, Tewodros Amberbir Habtegebrial and Didier Stricker    
     
  • Correlation of Pre-Operative Cancer Imaging Techniques with Post-Operative Gross and Microscopic Pathology Images
    Gabriel Reines March, Xiangyang Ju and Stephen Marshall    
     
  • Evaluating Quantized Convolutional Neural Networks for Embedded Systems
    Simon O'Keeffe and Rudi Villing    
     
  • Stitching Skin Images of Scars
    S. M. Iman Zolanvari and Rozenn Dahyot    
     
  • Characterisation of CMOS Image Sensor Performance in Low Light Automotive Applications
    Shane P. Gilroy, John O'Dwyer and Lucas C. Bortoleto    

11.20am - 12.40pm Oral Paper Session 6 - Chair: Tom Naughton

  • Classification of Alzheimer’s disease subjects from MRI using the principle of consensus segmentation
    Aymen Khlif and Max Mignotte    
     
  • Multi-Class U-Net for Segmentation of Non-Biometric Identifiers
    Tomislav Hrkac, Karla Brkić and Zoran Kalafatic    
     
  • A Restricted-Domain Dual Formulation for Two-Phase Image Segmentation
    Jack Spencer    
     
  • Spatio-temporal tube segmentation through a video metrics-based patch similarity measure
    Patricia Vitoria Carrera, Vadim Fedorov and Coloma Ballester    

12.40pm - 1pm Best Paper Awards & Closing Remarks