Who We Are
The AI in Multimedia – Image and Video Processing Lab (AIM-IVPL) at Northwestern University is currently composed of our lab director Aggelos Katsaggelos, postdoctoral researchers, doctoral students, and a handful of masters and undergraduate students. For more information, please check out our People section.
What We Do
Our work mainly focuses on solving problems in five fields and applications of research. Many of our projects lie at the intersection of these fields, and we pride ourselves in the way our research connects the areas together. The Venn diagram below is key to our identity as a lab, the research we conduct, and the projects we take on. To view our projects, please check out the Projects page.
Machine learning is everywhere!
Machine learning is a hot topic these days! Most of our work involves machine learning, deep learning, or other data science techniques to tackle the problems we face. Whether we design a new network to achieve the best results, or we use an existing network in a novel way, machine learning is a major part of the work we do. Our lab has exclusive access to a total of 15 GPUs for the purpose of training and evaluating neural networks.
To see what projects we’re working on involving this topic, please check out our Machine Learning Projects page.
Artificial intelligence in medicine will change how we treat patients
AI in medicine is an exciting front that has very real consequences in the world. Our lab is working to advance the field of medical imaging to ultimately improve the healthcare that patients receive. We have collaborated with doctors and medical researchers alike on projects such as brain lesion segmentation and early detection of Alzheimer’s disease.
To see what projects we’re working on involving this topic, please check out our AI in Medicine Projects page.
Signal processing is the basis of IVPL’s work
Our lab first started out in developing traditional/classical methods to address the image processing and analysis problems challenging researchers at the time. For example, image restoration techniques (the area where IVPL mainly pursued in the years after its inception) have come a long way since then, but is still recognizably the same field with the same goals. Today, we are still working on image restoration techniques, as well as other classic topics such as image and video super-resolution and object tracking.
To see what projects we’re working on involving this topic, please check out our Image Processing and Analysis page.
Cultural heritage is a vital field in the arts
Beyond the applications of data science we may typically think of, we love to apply our knowledge in the field of cultural heritage. More specifically, IVPL has teamed up with other research groups and institutions to study works of art. We are currently working towards developing algorithms to identify the pigments that artists used without damaging the painting, as well as the design of sampling and inpainting techniques of different imaging modalities in an effort to speed up the acquisition time.
To see what projects we’re working on involving this topic, please check out our Cultural Heritage page.
Computational imaging expands the capabilities of cameras
Computational imaging for leveraging novel optical designs and image processing together to enable new capabilities in conventional cameras, such as extended depth-of-field, digital refocusing, super-resolution, and measuring high dimensional appearance.
To see what projects we’re working on involving this topic, please check out our Computational Imaging Projects page.