sd-webui-modelscope-text2video
erasing
sd-webui-modelscope-text2video | erasing | |
---|---|---|
17 | 11 | |
479 | 475 | |
- | - | |
10.0 | 5.0 | |
about 1 year ago | 6 months ago | |
Python | Python | |
Apache License 2.0 | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
sd-webui-modelscope-text2video
-
New 1.2B parameter text to video model is out, higher quality than modelscope
Working on it https://github.com/deforum-art/sd-webui-modelscope-text2video/pull/96 (also, will rename the repo to just sd-webui-text2video after that)
- This is fine
-
Can someone help me understand what happens with VRAM?
They're linked from the main project's REDME.md under the "Where to get the weights" heading. (https://github.com/deforum-art/sd-webui-modelscope-text2video)
- Trump VS Godzilla - ModelScope + Img2Img
- I'm the creator of LoRA. How can I make it better?
-
"Melting World" - Text To Video
Workflow: Text to video AUTO1111 extension https://github.com/deforum-art/sd-webui-modelscope-text2video
-
Wake up, samurai! ModelScope text2video fine-tuning repo just dropped! Based on Diffusers, requirements start from GTX 3090 at the moment
Please, give it a try and leave your feedback. Soon fine-tuned models are planned to be usable in the Auto1111 plugin https://github.com/deforum-art/sd-webui-modelscope-text2video/issues/48 as well
-
ModelScope text2video is reported to be running at 4 GBs of VRAM with enough effort โ still, help needed to bring more optimizations and streamline the process
Meanwhile, if you have good training vids, it'd be nice to collect them somewhere for the future training, like inside the extension repo's Discussions https://github.com/deforum-art/sd-webui-modelscope-text2video/discussions
- The kind of result I'm getting with the new A1111 MS text2video model on a RTX 3060 (12 GB)
-
"The Rise Of AI" - Text To Video Short Film
Workflow: Text to video AUTO1111 extension https://github.com/deforum-art/sd-webui-modelscope-text2video
erasing
-
Unified Concept Editing in Diffusion Models (erase art styles in seconds )
Editing models in seconds. This is an upgrade to the previous style erasure https://erasing.baulab.info but faster training with no damage to the model prior knowledge!
-
Unified Concept Editing in Diffusion Models
Editing models in seconds. This is an upgrade to the lora sliders (https://erasing.baulab.info and https://github.com/p1atdev/LECO) but faster training with no damage to the model prior knowledge! Check out their code: https://github.com/rohitgandikota/unified-concept-editing
-
Using stable diffusion eraser to replace a concept in one model with the same concept from another
Figured I'd drop a link to the proof-of-concept. Ashen ran a color replace on my sanxianwonton mix a while ago. Sanxianwonton is a very monochromatic checkpoint because it is made to mimic Chinese shuimohua (ink wash) style painting. However by using the eraser tool from Rohit Gandikota it's possible to add color from another model, in this case colorBoxModel which sells itself as a model specifically made for color. Essentially the process is:
- Are there any sure-fire 100% SFW models for Stable Diffusion? Project for kids
-
ModelScope text2video test - prompt: "a skeleton dancing"
this might work to remove it: https://github.com/rohitgandikota/erasing or other watermark removing techniqueus
-
First open source text to video 1.7 billion parameter diffusion model is out
Wouldn't be surprising to see Automatic1111 to incopore it in A1111 web ui and add the eraser option for that f* logo. https://github.com/rohitgandikota/erasing
-
This AI Paper Proposes A New Method For Fine-Tuning Model Weights To Erase Concepts From Diffusion Models Using Their Own Knowledge
Quick Read: https://www.marktechpost.com/2023/03/17/this-ai-paper-proposes-a-new-method-for-fine-tuning-model-weights-to-erase-concepts-from-diffusion-models-using-their-own-knowledge/ Paper: https://arxiv.org/pdf/2303.07345.pdf Github: https://github.com/rohitgandikota/erasing
-
New research: Erasing concepts from diffusion models
Paper: https://arxiv.org/pdf/2303.07345.pdf
- Erasing Concepts from Stable Diffusion
What are some alternatives?
diffusers - ๐ค Diffusers: State-of-the-art diffusion models for image and audio generation in PyTorch and FLAX.
modelscope - ModelScope: bring the notion of Model-as-a-Service to life.
sd-webui-additional-networks
LECO - Low-rank adaptation for Erasing COncepts from diffusion models.
VideoCrafter - VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models
sd_lite - set-up Stable Diffusion with minimal dependencies and a single multi-function pipe
Text-To-Video-Finetuning - Finetune ModelScope's Text To Video model using Diffusers ๐งจ
sd-webui-text2video - Auto1111 extension implementing text2video diffusion models (like ModelScope or VideoCrafter) using only Auto1111 webui dependencies
lora - Using Low-rank adaptation to quickly fine-tune diffusion models.
ebsynth_utility - AUTOMATIC1111 UI extension for creating videos using img2img and ebsynth.
ComfyUI - The most powerful and modular stable diffusion GUI, api and backend with a graph/nodes interface.