Timezone: »

Flexible Diffusion Modeling of Long Videos
William Harvey · Saeid Naderiparizi · Vaden Masrani · Christian Weilbach · Frank Wood

Wed Nov 30 09:00 AM -- 11:00 AM (PST) @ Hall J #514

We present a framework for video modeling based on denoising diffusion probabilistic models that produces long-duration video completions in a variety of realistic environments. We introduce a generative model that can at test-time sample any arbitrary subset of video frames conditioned on any other subset and present an architecture adapted for this purpose. Doing so allows us to efficiently compare and optimize a variety of schedules for the order in which frames in a long video are sampled and use selective sparse and long-range conditioning on previously sampled frames. We demonstrate improved video modeling over prior work on a number of datasets and sample temporally coherent videos over 25 minutes in length. We additionally release a new video modeling dataset and semantically meaningful metrics based on videos generated in the CARLA autonomous driving simulator.

Author Information

William Harvey (University of British Columbia)
Saeid Naderiparizi (University of British Columbia)
Vaden Masrani (University of British Columbia)
Christian Weilbach (University of British Columbia)
Frank Wood (University of British Columbia)

More from the Same Authors