Join us Online for LIM2021: 20-21 September
About the London Imaging Meeting
The London Imaging Meeting (LIM) is a yearly topic-based conference focused on the future of imaging science. The theme for 2021 is “Imaging for Deep Learning”. Proposals for the 2022 theme are being accepted (see below).
The 2021 conference is anticipated to take place over 3 days, with the technical
program held online 20-21 September, plus an additional day for
short courses. Details on courses will be released later.
On this page
LIM2021: CALL FOR PAPERS
The theme of LIM 2021 is IMAGING FOR DEEP LEARNING. We are interested in two aspects of deep-learning and imaging science:
- Imaging science to improve deep learning performance
- Deep learning methods intended to improve imaging systems performance
We seek to attract new research in areas relating to the ways imaging system design, processes, their performance, and outputs affect deep neural network (DNN) performance.
We are also interested in deep learning methods to improve imaging: how deep neural networks are used to optimize imaging system design, predict performance, and output image quality.
The conference is designed to promote interaction between the imaging science and deep learning communities. It includes research, keynote, and focus presentations.
LIM 2021 PROGRAM TOPICS
Topics of interest include, but are not limited to:
- Impact of imaging system design on DNN performance
- Imaging system optimization for improved DNN performance
- Imaging system modelling for data augmentation for DNN training
- Metrification of imaging performance for DNN input
- DNNs for metrification of imaging systems
- DNNs for optimization of imaging performance
- DNNs for image quality prediction
- DNN components for imaging systems
- Domain adaptation of DNNs for imaging systems
In their papers, we encourage authors to speculate on how their research in imaging science will impact the field of deep learning in the future and vice versa. We welcome participation of scientists and engineers from academia and industry, and strongly encourage contributions from graduate students. LEARN MORE
LIM 2021 KEYNOTE PRESENTATIONS
We are pleased to announce our two distinguished LIM 2021 keynote speakers.
Soft-Prototyping Camera Designs for Autonomous Driving
Dr. Joyce E. Farrell, executive director
Stanford University Center for Image Systems Engineering (SCIEN)
view abstract
Camera Metrics for Autonomous Vision (working title)
Dr. Robin Jenkin, principal image quality engineer
NVIDIA
FOCAL TALKS
Deep learning in image quality assessment: past, present, and what lies ahead
Seyed Ali Amirshahi,
associate professor, Norwegian University of Science and Technology (NTNU)
Image understanding for color constancy and vice versa
Simone Bianco,
associate professor of Computer Science, Università degli Studi di Milano-Bicocca
The data conundrum: compression of automotive imaging data and deep neural network based perception
Valentina Donzella,
associate professor, Intelligent Vehicles Group, University of Warwick
Mitigating limitations of deep neural networks for imaging systems
Ray Ptucha,
computational display technology leader, Apple
Generative models for image data augmentation
Jonas Unger,
professor, Linköping University
Online 2021
LIM2021 has moved online. The meeting's online logistics are being organized and run by the Institute of Physics (IOP). As details become available, they will be posted here.
LIM 2022
The LIM Steering Committee seeks inquiries from colleagues/groups interested in proposing a theme and/or chairmanship for the 2022 London Imaging Meeting. Inquiries/proposals may be directed to
london@imaging.org.
Funding for the conference keynotes and proceedings is supported by EPSRC.