Anthony Fuller | Three Methods for Efficient Computer Vision

Video Description

Anthony's presentation introduces three groundbreaking methods for efficient computer vision applications. His research includes Jumbo, which enhances vision transformers (ViTs) with global information processing at minimal computational cost; LookWhere, which makes high-resolution computer vision accessible by intelligently processing only image regions of interest through "where" and "what" computational division; and Self-souping, which creates effective model soups by averaging multiple model weights without requiring labels through self-supervised learning techniques. Internship Opportunities at Vector Institute Advance your career in AI with Vector's internship program. Collaborate with leading AI developers on cutting-edge projects, gain invaluable insights through daily research talks, and benefit from professional development opportunities. Vector offers diverse internship tracks including Applied AI, Research, and specialized programs for Black and Indigenous students. Enhance your professional growth through Vector's Mentorship Program, pairing you with an AI expert from a sponsor organization to strengthen your technical and interpersonal skills. Take the next step in your AI journey. Learn more at: https://vectorinstitute.ai/programs/irap-bias-in-ai/ About Vector: The Vector Institute drives excellence and leadership in Canada's knowledge, creation, and use of artificial intelligence to foster economic growth and improve the lives of Canadians. Through research, industry collaboration, and talent development, Vector advances AI innovation across sectors while building Canada's leadership in the field. Remarkable 2026 is coming, subscribe to our newsletter to learn more: https://vectorinstitute.ai/#newsletter Connect with Vector: https://vectorinstitute.ai/ https://x.com/VectorInst https://www.linkedin.com/company/vector-institute