Ph.D. student Anna Fruehstueck wins 2020 Facebook Fellowship Award

Anna Fruehstueck, a KAUST Ph.D. student in computer science, recently won a 2020 Facebook Fellowship award and a two-year fellowship from Facebook Research. Photo by Khulud Muath.

​-By David Murphy, KAUST News

KAUST Ph.D. student Anna Fruehstueck recently won a 2020 Facebook Fellowship award and a two-year fellowship from Facebook Research. Fruehstueck is part of the University's Visual Computing Center and is supervised by Professor Peter Wonka. 

The Facebook Fellowship supports and encourages a global base of doctoral students who are active in innovative and relevant research related to computer science and engineering. Fruehstueck's fellowship focuses on computer graphics and will begin in Fall 2020. She was was selected as one of 36 recipients of the award from an initial pool of 1,800 applicants and is the University's first recipient of the scholarship.

"Facebook Research only selected four recipients in the area of computer graphics," Fruehstueck stated. "The fellowship is very competitive, and the 36 recipients for 2020 were picked from 1,876 applicants from over 100 universities worldwide. Additionally, the 2020 recipients were selected from 16 universities in the U.S. and only six universities outside the U.S."

"The fellowship benefits include a financial award and an invitation to meet other fellows and the Facebook Research team at a Summit Event at the Facebook Headquarters," she continued. "The fellowship will run for two years or until I graduate, whichever comes first. I am really excited to connect with the computer graphics research department at Facebook Research."

Ph.D. student Anna Fruehstueck's Facebook Fellowship includes a financial award and an invitation to meet the Facebook Research team and other fellows at Facebook Headquarters. Image courtesy of Shutterstock.

The promise of new possibilities

Fruehstueck obtained her bachelor's degree in computer science and her master's degree in visual computing from the Vienna University of Technology, Austria, in 2011 and 2015, respectively. During her master's degree, Fruehstueck came to the University as a visiting student and experienced the environment and spirit of learning and innovation on campus. 

"Before I joined KAUST, I was unsure if academia was a path that I wanted to follow," she said. "Visiting KAUST for the first time opened my eyes to new possibilities, as I was immediately taken by how passionate and excited people are about their work. I am delighted to be working at KAUST, and I think the environment here and being surrounded by motivated, productive and talented people is a critical component in my creative research process."

Fruehstueck's research at KAUST explores the intersection of computer graphics, geometry processing and visual arts using state-of-the-art methods in machine learning. She focuses on the development of algorithms for image synthesis and analysis and the design of generative modeling techniques. 

She currently works on designing algorithms that can generate novel high-resolution images based on a collection of input images—for example, generating satellite images of a fictional landscape by learning from preexisting aerial image data or generating a fictional artwork by emulating the brush strokes of an existing painting.

"My concerned with the synthesis of natural texture images, which are an important component in many applications ranging from architectural rendering to game design and often require the manual work of skilled artists," she explained.

"Our work—titled "TileGAN: Synthesis of Large-Scale Non-Homogeneous Textures"—was published as a technical paper at ACM SIGGRAPH 2019, which is the most prestigious conference in computer graphics, and I presented [the] paper at the conference in July 2019 in Los Angeles," Fruehstueck said.

Anna Fruehstueck and her KAUST Visual Computing Center colleagues generated the painting pictured of Jeddah's Al-Balad historic area using image synthesis techniques in the style of famous painter Georges Seurat. Image courtesy of Anna Fruehstueck.

Future work

Fruehstueck's future research will continue focusing on creating new techniques in image synthesis to allow for the generation of higher-quality images and better output image control.

"Given the ubiquity of image data in today's life, I believe my research is relevant for a large variety of fields, and I hope that my work will contribute to tools that are relevant for real-life applications," she noted. "This year, I also have the opportunity of doing internships with two companies, first working at Weta Digital and then at Adobe. I am excited about these collaborations and hope to gain valuable insights into other research groups, and I look forward to meeting new collaborators and learn[ing] from them."

Related stories: