Day 1 - Special Sessions Schedule

09:10 – 09:20: Workshop Intro
Workshop organisers

09:20 – 09:45: Keynote 1: Recent Advancements in Perception & Planning
Dr. Andrei Bursuc, Deputy Scientific Director at valeo.ai, Paris, France

09:45 – 10:10: Keynote 2: Generative AI in Autonomous Driving
Dr. Ibrahim SOBH, Senior Expert in AI, Cairo, Egypt, LinkedIn Top Voice in AI

10:10 – 10:25: Paper 1: Deformable Convolution Based Road Scene Semantic Segmentation of Fisheye Images in Autonomous
Driving
Anam Manzoor,  Aryan Singh, Reenu Mohandas, Eoin Grua, Tony Scanlan, Ganesh Sistu, Ciarán Eising

10:25 - 10:40: Paper 2: Unlocking Immersive Driving Experiences with Automotive Sensor-Powered Extended Reality (XR)
Anushalakshmi Manila, Sonam Negi, Christopher Nowakowski, Patrice Reilhac

10:40 – 10:55: Paper 3: Velocity Driven Vision: Asynchronous Sensor Fusion Birds Eye View Models for Autonomous Vehicles
Seamie Hayes, Sushil Sharma, Ciarán Eising

10:55 – 11:10: Paper 4: Enhanced Parking Perception by Multi-Task Fisheye Cross-view Transformers
Antonyo Musabini, Ivan Novikov, Sana Soula, Christel Leonet, Lihao Wang, Rachid Benmokhtar, Fabian Bürger,
Thomas Boulay, Xavier Perrotton

Brian Deegan

Part 1 (60 minutes)

• Overview of camera technology

  • Optics

  • Image Sensors

  • Image Processing

• Introduction to Image Quality

  • Subjective vs objective image quality KPIs

  • Quality KPIs for deep learning/computer vision

  • Relevant standards (e.g. IEEE P2020, ISO12233, EMVA1288)

Part 2 (60 minutes)

• Relating image quality KPIs to algorithm performance

  • A review of the state of the art/literature

  • Camera mass production - variation in real-world camera performance

  • How to use image quality KPIs to design camera systems

• Challenges of existing databases and model training approaches

  • Limitations of public datasets

  • Simulated data considerations

  • Dataset generation - pitfalls to avoid

• On a positive note:

  • How to apply knowledge of camera technology, image quality, and tips for dataset generation to improve system performance

11:40 – 11:45: Workshop Intro
Workshop organisers

11:45 – 12:20: Workshop Keynote:  Explainable AI for Clinical Decision Support Systems
Dr Ayse Keles, University of Galway

12:20 – 12:40: Paper 1:  Comparative Analysis of Real-Clinical MRI and BraTS Datasets for Brain Tumor Segmentation
Ramin Ranjbarzadeh, Ayse Keles, Martin Crane, Malika Bendechache

12:40 – 13:00: Paper 2:  Improving Diagnostic Trust: An Explainable Deep Learning Framework for Genitourinary Cancer
Prediction
Dost Muhammad, Rafi Ullah, Malika Bendechache

13:00 – 13:20: Paper 3: Subgraph Clustering and Atom Learning for Improved Image Classification
Aryan Singh, Pepijn van de Ven, Ciarán Eising, Patrick Denny

13:20 – 13:40: Paper 4: Assessment of Skin Type Measurement in Image Datasets
Neda Alipour, Ted Burke, Jane Courtney

11:40 – 12:20: Workshop Intro and Keynote
Prof Peter Corcoran, University of Galway

12:20 – 12:40: Paper 1:  Spiking-DD: Neuromorphic Event based Spiking Driver Distraction Detection
Waseem Shariff, Paul Kielty, Joseph Lemley, Peter M Corcoran

12:40 – 13:00: Paper 2:  MapsTP: HD Map Images Based Multimodal Trajectory
Prediction for Automated Vehicles
Sushil Sharma, Arindam Das, Ganesh Sistu, Mark Halton, Ciarán Eising

13:00 – 13:20: Paper 3: A Framework for Pupil Tracking with Event Cameras
Khadija Iddrisu, Waseem Shariff, Suzanne Little

13:20 – 13:40: Paper 4: On Improving Solar Forecasts with Enhanced IR Imaging and CNNs
Ifran Rahman Nijhum, Md Yearat Hossain, Md Mahbub Hasan Rakib, Soumyabrata Dev

14:40 – 14:45: Workshop Intro
Workshop organisers

14:45 – 15:10: Workshop Keynote:  GANs to Diffusion models
Dr Muhammad Ali Farooq, University of Galway

15:10 – 15:30: Paper 1:  Handling occlusions via occlusion-aware labels
Fatemeh Amerehi, Patrick Healy

15:30 – 15:50: Paper 2:  SS-SFR: Synthetic Scenes Spatial Frequency Response on Virtual KITTI and Degraded Automotive
Simulations for Object Detection
Daniel Jeno Jakab, Alexander Braun, Cathaoir Agnew, Reenu Mohandas, Brian Deegan, Dara Molloy, Enda Ward,
Tony Scanlan, Ciarán Eising

15:50 – 16:10: Paper 3: A segmentation method for synthetically-trained pose-based fingerspelling recognition models
Frank Fowley, Anthony Ventresque

16:10 – 16:30: Paper 4: Advancing Turfgrass Maintenance with Synthetic Data for Divot Detection
Stephen Foy, Simon D Mcloughlin

16:30 - 16:40: Short paper lightning talk: Investigating the Impact of Encoder Architectures and Batch Size on Depth Estimation through Semantic Consistency
Nosheen, Iqra, Iqbal, Talha, Ullah, Ihsan, Ennis, Cathy, Madden, Michael G

Waqar Shahid Qureshi

The session will cover:

• • Fundamental principles and significance of SfM and MVS in computer vision, robotics, and augmented reality.

  • Introduction to openMVG's functionalities for SfM, encompassing feature detection, matching, and camera pose estimation, alongside an overview of openMVS for dense reconstruction, meshing, and texturing.

  • Installation guidance for openMVG and openMVS on different platforms, including setting up essential dependencies and libraries.

  • Tips for capturing images suitable for successful reconstruction, covering aspects like camera calibration, image quality, and overlap.

  • Practical demonstrations using openMVG for SfM, including feature detection, matching, incremental Structure from Motion, and global bundle adjustment.

  • Progressing to openMVS for dense reconstruction, covering mesh reconstruction, refinement, and texturing.

  • Techniques for refining and optimizing the resultant 3D model, followed by visualization using open-source tools.