Video Understanding: Human Level Performance Using Multimodal Foundation Models with a Video First Ethos

Explore video foundation models that process video holistically, mirroring human understanding. This talk presents a departure from LLMs, showcasing human-level video comprehension.

Overview

Video has traditionally be processed as its constituents: frames and audio but this isn’t how humans process video. Learn how video foundation models take a video first approach to enabling human level understand of video.

Links

Tech stack