Blog Seminars and Talks Vacancies Job Postings Activities Datasets Theses ... Seminars and Talks ... In this talk, I will introduce some recent works that explore learning from appearance, motion and language in an unsupervised or weakly-supervised manner.
Blog Seminars and Talks Vacancies Job Postings Activities Datasets Theses ... In this talk, we will focus on Polynomial Networks (PNs), which use high-degree polynomial expansions to approximate the target function. ... In the second part of the talk, we will extend PNs for conditional tasks where we have multiple (possibly diverse) inputs.
Blog Seminars and Talks Vacancies Job Postings Activities Datasets Theses ... Seminars and Talks ... How can we infer and represent such regular structures from raw visual data, and later exploit them for better scene recognition, synthesis, and editing? In this talk, I will present our recent work on developing neuro-symbolic methods for scene understanding.
Blog Seminars and Talks Vacancies Job Postings Activities Datasets Theses ... Seminars and Talks ... In this talk, I will cover our two recent works on reformulating DDMs specifically for fast sampling.
Blog Seminars and Talks Vacancies Job Postings Activities Datasets Theses ... Seminars and Talks ... In this talk, I will discuss recent works by the NVIDIA Toronto AI Lab on diffusion models.
Blog Seminars and Talks Vacancies Job Postings Activities Datasets Theses ... Seminars and Talks ... In this talk, I will share some of my recent work in building better video representations, by leveraging these advanced representations, specifically for the tasks of video understanding, multimodal understanding, ...
Blog Seminars and Talks Vacancies Job Postings Activities Datasets Theses ... Seminars and Talks ... In this talk, we present new neural architectures as well as learning and inference frameworks to understand complex activity videos, addressing the following challenges: ...
Blog Seminars and Talks Vacancies Job Postings Activities Datasets Theses ... Seminars and Talks ... In this talk, we will study the internal representations of two prominent foundation models: DINO-ViT - a self-supervised vision transformer, and StableDiffusion - a text-to-image generative latent diffusion model.
Blog Seminars and Talks Vacancies Job Postings Activities Datasets Theses ... In this talk, we show how insights from unnormalized statistical modeling and convex optimization can overcome these limitations. ... We will conclude the talk with some ideas for future work.
Blog Seminars and Talks Vacancies Job Postings Activities Datasets Theses ... All talks ... Seminars and Talks ...