In the 1995 movie “Clueless,” lead character Cher Horowitz has a digital closet that allows her to virtually try on outfits. Proving some ideas never go out of style, Allen School professor Ira Kemelmacher-Shlizerman has spent the past two decades working to make that and other futuristic technologies a reality.
“My general area is at the intersection of computer vision and graphics, or generative media. I am excited about all aspects of image and video generation and the applications we can build on top of it. A big passion of mine is to model people and clothing from large photo collections,” said Kemelmacher-Shlizerman, director of the UW Reality Lab and member of the UW Graphics & Imaging Group (GRAIL). “I am currently working to push the boundaries of human and clothing modeling to new levels, going to extreme quality and details.”
The Institute of Electrical and Electronics Engineers (IEEE) recently recognized her in the 2026 class of IEEE Fellows, one of the organization’s highest honors, for her “contributions to face, body, and clothing modeling from large image collections.” The IEEE Fellows represent members with an exceptional record of accomplishments in their field who bring the “the realization of significant value to society at large.”
Kemelmacher-Shlizerman pioneered the field of face modeling from large photo collections in the wild. She helped develop the first photometric stereo method that can reconstruct three-dimensional face models from unstructured photos pulled from the Internet, launching thousands of follow-up papers. As the field progressed, Kemelmacher-Shlizerman and her collaborators introduced the MegaFace Benchmark, the first million-image scale image dataset for evaluating facial recognition algorithms. The work served as the field’s standard benchmark for many years. Kemelmacher-Shlizerman then developed a personalized image search engine that allows users to imagine how they could look with various hairstyles, or in different time periods or ages — anything that can be queried in the search engine. She later commercialized the technology through her startup Dreambit, which was acquired by Meta.
Since then, Kemelmacher-Shlizerman has transitioned from helping users test out new hairstyles to trailblazing virtual try-on technology, allowing for real clothing to be rendered on a human body. She and her team introduced the first use of conditional generative adversarial networks for photorealistic try-on. In addition to virtual try-on technology for photos, Kemelmacher-Shlizerman developed Fashion-VDM, a video diffusion model for generating virtual try-on videos to help users see the garment from multiple angles and understand how it flows and drapes in motion.
“We are working on detailed measurement of humans and clothing to enable fit aware virtual try-on, as in going beyond generative visualization to providing metrically correct measurement based try-on to create better than physical try-on technology,” said Kemelmacher-Shlizerman.
Her work as a principal scientist at Google, where she leads Google Shopping’s Generative Media team, is bringing this technology to the mainstream. Kemelmacher-Shlizerman and her team introduced a generative AI tool for Google Shopping where users can see how an article of clothing looks on a range of models of different sizes, body shapes and skin tones. Last December, her team launched an upgraded version of the virtual try-on tool that generates a full body digital version of a user to help them see how clothes would look like on their own body.
“Ira is arguably the foremost researcher in the field of virtual try-on technology. She has written several seminal papers, pioneering the use of generative AI for this use case, and is the chief technologist behind Google’s launch of this technology in their search and shopping products,” said Allen School professor Steve Seitz, who co-directs the UW Reality Lab. “While this research area is relatively new, it’s already having a major impact on industry.”
In addition to being named an IEEE fellow, Kemelmacher-Shlizerman has been recognized as an Association for Computing Machinery (ACM) Distinguished Member and received a Google Faculty Award, GeekWire Innovation of the Year Award, and the Madrona Prize.
Read more about the IEEE Fellow Class of 2026.