Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

updates for animation support #116

Draft
wants to merge 238 commits into
base: main
Choose a base branch
from
Draft

updates for animation support #116

wants to merge 238 commits into from

Conversation

dmarx
Copy link
Contributor

@dmarx dmarx commented Nov 2, 2022

No description provided.

dmarx and others added 30 commits October 30, 2022 13:39
* add `interfaces/__init__.py` via CI
* dump unnecessary file copy stuff
* flushed vestigial files
* cleanup overengineered opencv CI
- download and import stability-sdk anima branch
- checkout specific commit so sdk can change without breaking notebook
- start migrating to utility functions in SDK
- color_coherence options for HSV, RGB, and LAB
- tweaked default values
- new anim curves for: cadence, fov, video_mix_in
- move cam settings to bottom of settings, add in animation_prompts and negative prompt
- skip mp4 for run all
kostarion and others added 29 commits April 26, 2023 00:15
Mask fixup is now done by default on the generator_server side
- inpainting model performs full steps
- non-inpainting model start_schedule and steps depend on the mask
- mask_pow in mask postprocessing
- mask erosion and blur with the same kernel
- remove non_inpainting_model_for_diffusion_frames param
- add mask_power param for 3D render mode mask postprocessing
- add use_inpainting_model param
- by default inpainting is done using non-inpainting models
- mask postprocessing inside inpaint function
- seed is not fixed for inpainting with non-inpainting model
- 3D render mode adjustments to new depth
- Outpaint with inpainting model
- New preset Translate and inpaint
* Dynamic transform routing to CPU node
* cv2 -> PIL to match main SDK and ease the transition StabilityInference -> api.Context
* Eliminate issue of shared utils.py bringing in dependencies only available with [anim] install.
* Update README.md with animation command
* Update animation tests and add new tests for each transform method
* Upscale method on api.Context, upscale 2X option in Post-process tab of animation_ui
* Display callstacks with exceptions in animation_ui
* Remove warning about new API being in beta
* Tweaks
  - update default animation sampler
  - change steps_strength_adj default to False
  - validation on border modes
  - find latest revision of project file
  - handle exceptions while getting user credits and updating header during animation rendering
* Clean up utils.py
  - convert all the _from_string functions to use shared table lookup
  - group and alphabetize
  - add doc strings to the general utility functions
* Graceful handling of STABILITY_KEY not being set.
* Merge latest from main
- Fix span/outpaint which got borked by move to PIL
- Fix Project import and delete and give Confirm button before deleting
- Handle exceptions from create_video_from_frames
- Update clip_guidance default to None to match docs
- Update DATA_GENERATOR name that goes in saved settings files
- Give example prompt for "3D warp rotate" preset and use pure MiDaS for better result
…ess_video

Cut pow/resume and postprocess video
avoid 'prefill' border when 'inpaint_border' is disabled
* Provide user more feedback when stopping anim/post and while creating mp4 from frames.
* Ensure animation dimensions are a multiple of 64
* Switch from Colab form UI to `getpass` for entering API keys.
@sonarqubecloud
Copy link

Kudos, SonarCloud Quality Gate passed!    Quality Gate passed

Bug A 0 Bugs
Vulnerability A 0 Vulnerabilities
Security Hotspot A 0 Security Hotspots
Code Smell A 0 Code Smells

No Coverage information No Coverage information
0.0% 0.0% Duplication

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants