Get Started
Home
Topics
Search
Library
Research questionHow can text-to-video generation continuously edit local attributes while preserving unrelated content?Changing one visual attribute can unintentionally alter other parts of a generated video. The problem is especially difficult when suitable examples of corresponding edited videos are unavailable.
Computer Vision
Diffusion Models
Image & Video Processing
Machine Learning
Video Generation
Latest papersRecent research connected to this question, newest first.TokenDial: Continuous Attribute Control for Text-to-Video Generation in Visual Dial SpaceThe source learns reusable additive controls in a frozen video diffusion generator from the effects they induce on generated videos rather than paired edited-video supervision. Experiments and human studies examine localization, composition, and reuse across prompts, resolutions, and video lengths.research paper · Sep 4, 2026
Related questions
How can language-controlled video generators make character and camera actions temporally precise in interactive worlds?How can causal streaming video editing remain real-time while preserving backgrounds and unedited regions over long sequences?How can video editing handle diverse instruction- and subject-guided edits while preserving coherence and identity?How can reference-guided image generation control distinct attributes of multiple objects in complex scenes?