Skip to content

yashaswip/Text-2-Video-generation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

34 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Tune-A-video:

Final results from the evaluation of the Tune-A-Video model reveal an intriguing outcome. The model was trained on videos of a duck diving underwater in search of food from right to left. However, when prompted with "A fish diving underwater in search of food from right to left," the model successfully generated a video depicting the desired underwater sequence featuring a fish.

The generated video demonstrated strong temporal coherence, with a high frame consistency score of 0.9335966. This indicates that the model effectively maintained visual consistency throughout the sequence, even though it was originally trained on a different animal. The smooth transitions and dynamic motion of the fish highlight the model's ability to adapt its learned behaviors to new prompts, capturing intricate underwater actions with notable detail and fluidity.

This outcome showcases the model's versatility and its potential for generating realistic video sequences across different subjects, demonstrating its capability to generalize from training data to varied prompts and enhancing its utility in AI-driven video generation technology..

Unknown-3 A Duck diving underwater in search of food from right to left

Unknown A Fish diving underwater in search of food from right to left

Motion Director

When generating videos based on prompts, such as "A dolphin in ocean zoom in," showcasing the model's ability to produce realistic and dynamic video sequences from textual descriptions. Unknown-5

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published