Instant Video Models: Universal Adapters for Stabilizing Image-Based Networks
Matthew Dutson, Nathan Labiosa, Yin Li, Mohit Gupta · 2025
Read Instant Video Models: Universal Adapters for Stabilizing Image-Based Networks by Matthew Dutson, Nathan Labiosa, Yin Li, Mohit Gupta (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-03014.txt.
Description
Matthew Dutson and three co-authors adapt still image models for video without letting outputs flicker. Frame based networks applied one frame at a time often show temporal inconsistency. Time varying corruptions in the input make the flicker worse, as each frame is cleaned differently.
Their Instant Video Models act as universal adapters for image based networks. The adapters stabilize inference and add robustness across the video. One adapter for many models saves retraining each network for video, a gain for restoration, enhancement, and analysis tools that now work frame by frame. One wrap for many tools is a thrift that counts. Each tool would else need weeks of new work to run on film at a fair rate of speed.
Catalog record
| CREATOR | Matthew Dutson, Nathan Labiosa, Yin Li, Mohit Gupta |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 82 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-03014