Skip to content

Releases: volotat/SD-CN-Animation

v0.9

14 May 03:00
Compare
Choose a tag to compare

Last version changes: v0.9

  • Fixed issues #69, #76, #91, #92.
  • Fixed an issue in vid2vid mode when an occlusion mask computed from the optical flow may include unnecessary parts (where flow is non-zero).
  • Added 'Extra params' in vid2vid mode for more fine-grain controls of the processing pipeline.
  • Better default parameters set for vid2vid pipeline.
  • In txt2vid mode after the first frame is generated the seed is now automatically set to -1 to prevent blurring issues.
  • Added an option to save resulting frames into a folder alongside the video.
  • Added ability to export current parameters in a human readable form as a json.
  • Interpolation mode in the flow-applying stage is set to ‘nearest’ to reduce overtime image blurring.
  • Added ControlNet to txt2vid mode as well as fixing #86 issue, thanks to @mariaWitch
  • Fixed a major issue when ConrtolNet used wrong input images. Because of this vid2vid results were way worse than they should be.
  • Text to video mode now supports video as a guidance for ControlNet. It allows to create much stronger video stylizations.

v0.8

11 May 04:35
Compare
Choose a tag to compare
  • Better error handling. Fixes an issue when errors may not appear in the console.
  • Fixed an issue with deprecated variables. Should be a resolution of running the extension on other webui forks.
  • Slight improvements in vid2vid processing pipeline.
  • Video preview added to the UI. It will become available at the end of the processing.
  • Time elapsed/left indication added.
  • Fixed an issue with color drifting on some models.
  • Sampler type and sampling steps settings added to text2video mode.
  • Added automatic resizing before processing with RAFT and FloweR models.

v0.7

07 May 19:37
Compare
Choose a tag to compare
  • Text to Video mode added to the extension
  • 'Generate' button is now automatically disabled while the video is generated
  • Added 'Interrupt' button that allows to stop video generation process
  • Now all necessary models are automatically downloaded. No need for manual preparation.

v0.6

05 May 03:00
Compare
Choose a tag to compare
  • Complete rewrite of the project to make it possible to install as an Automatic1111/Web-ui extension.
  • Added flow normalization before resizing it, so the magnitude of the flow computed correctly at the different resolution.
  • Less ghosting and color drift in vid2vid mode
  • Added "warped styled frame fix" at vid2vid mode that removes duplicates from the parts of the image that cannot be relocated from the optical flow.

v0.5

19 Apr 00:36
Compare
Choose a tag to compare
  • Fixed an issue with the wrong direction of an optical flow applied to an image.
  • Added text to video mode within txt2vid.py script. Make sure to update new dependencies for this script to work!
  • Added a threshold for an optical flow before processing the frame to remove white noise that might appear, as it was suggested by @alexfredo.
  • Background removal at flow computation stage implemented by @CaptnSeraph, it should reduce ghosting effect in most of the videos processed with vid2vid script.

v0.4

12 Apr 21:23
7f32fc1
Compare
Choose a tag to compare
  • Fixed issue with extreme blur accumulating at the static parts of the video.
  • The order of processing was changed to achieve the best quality at different domains.
  • Optical flow computation isolated into a separate script for better GPU memory management. Check out the instruction for a new processing pipeline.

v0.3

08 Apr 12:59
Compare
Choose a tag to compare
  • Flow estimation algorithm is updated to RAFT method.
  • Difference map now computed as per-pixel maximum of warped first and second frame of the original video and occlusion map that is computed from forward and backward flow estimation.
  • Added keyframe detection that illuminates ghosting artifacts between the scenes.

v0.2

08 Apr 06:24
68bd68d
Compare
Choose a tag to compare

Initial release.

  • The Gunnar-Farneback optical flow estimation added.
  • Double pass for SD processing added. It allows to gather information from the current frame, as well as warped previous frame.
  • ControlNet API call updated to a new version.