Towards Long-Form Video Understanding
This addresses the challenge of contextualizing visual events over extended periods for applications in video analysis and AI systems, representing a domain-specific advancement.
The paper tackles the problem of long-form video understanding, showing that existing short-term models are limited for such tasks, and introduces a novel object-centric transformer-based architecture that significantly outperforms state-of-the-art methods on 7 diverse tasks and the AVA dataset.
Our world offers a never-ending stream of visual stimuli, yet today's vision systems only accurately recognize patterns within a few seconds. These systems understand the present, but fail to contextualize it in past or future events. In this paper, we study long-form video understanding. We introduce a framework for modeling long-form videos and develop evaluation protocols on large-scale datasets. We show that existing state-of-the-art short-term models are limited for long-form tasks. A novel object-centric transformer-based video recognition architecture performs significantly better on 7 diverse tasks. It also outperforms comparable state-of-the-art on the AVA dataset.