limuloo / migc Goto Github PK
View Code? Open in Web Editor NEW[CVPR 2024 Highlight] "MIGC: Multi-Instance Generation Controller for Text-to-Image Synthesis" (Official Implementation)
License: Other
[CVPR 2024 Highlight] "MIGC: Multi-Instance Generation Controller for Text-to-Image Synthesis" (Official Implementation)
License: Other
When I used MIGC to generate "2 cats and 3 dogs", I found that the first dog below would still look like a "cat". Is there any way to improve this result?
I am using realisticVisionV51_v51VAE.safetensors, here are my parameters:
rompt_final = [['4k, best quality, masterpiece, ultra high res, ultra detailed,a cat,a cat,a dog,a dog,a dog,grass',
'a cat', 'a cat', 'a dog', 'a dog', 'a dog', 'grass']]
bboxes = [[[0.078125, 0.09375, 0.390625, 0.359375], [0.515625, 0.09375, 0.859375, 0.359375], [0.078125, 0.515625, 0.34375, 0.90625],
[0.421875, 0.515625, 0.671875, 0.921875], [0.71875, 0.484375, 0.953125, 0.921875], [0.015625, 0.015625, 0.984375, 0.96875]]]
negative_prompt = 'worst quality, low quality, watermark, text, blurry'
seed = 12573842233801288171
seed_everything(seed)
image = pipe(prompt_final, bboxes, num_inference_steps=50, guidance_scale=8,
MIGCsteps=25, NaiveFuserSteps=25, aug_phase_with_and=False, negative_prompt=negative_prompt).images[0]
And here are the images generated by MIGC:
I'm here to ask for an update, when will the training code be released? May is almost over
I tried to generate a single image on my PC, it takes about 7 min on the single RTX 2060, is this normal?
Hi, thank you very much for your great work!
But I am wondering when MIGC supported with other SD models will be released?
I believe SD2.1 and SDXL can have a large boost compared to SD1.4.
Looking forward to your reply, thanks!
Thanks for the great work! However, the pipeline does not seem to support the latest diffusers. When I use diffusers v0.25.0, I got:
File "/home_dir/miniconda3/envs/env_name/lib/python3.10/concurrent/futures/_base.py", line 403, in __get_result
raise self._exception
File "/home_dir/miniconda3/envs/env_name/lib/python3.10/concurrent/futures/thread.py", line 58, in run
result = self.fn(*self.args, **self.kwargs)
File "/home_dir/MIGC/migc_gui/app.py", line 82, in process_request
pipe = offlinePipelineSetupWithSafeTensor(sd_safetensors_path=sd_safetensors_path)
File "/home_dir/MIGC/migc/migc_utils.py", line 174, in offlinePipelineSetupWithSafeTensor
pipe = StableDiffusionMIGCPipeline.from_single_file(sd_safetensors_path,
File "/home_dir/miniconda3/envs/env_name/lib/python3.10/site-packages/huggingface_hub/utils/_validators.py", line 118, in _inner_fn
return fn(*args, **kwargs)
File "/home_dir/miniconda3/envs/env_name/lib/python3.10/site-packages/diffusers/loaders/single_file.py", line 263, in from_single_file
pipe = download_from_original_stable_diffusion_ckpt(
File "/home_dir/miniconda3/envs/env_name/lib/python3.10/site-packages/diffusers/pipelines/stable_diffusion/convert_from_ckpt.py", line 1687, in download_from_original_stable_diffusion_ckpt
pipe = pipeline_class(
File "/home_dir/MIGC/migc/migc_pipeline.py", line 241, in __init__
super().__init__(
File "/home_dir/miniconda3/envs/env_name/lib/python3.10/site-packages/diffusers/pipelines/stable_diffusion/pipeline_stable_diffusion.py", line 237, in __init__
self.register_modules(
File "/home_dir/miniconda3/envs/env_name/lib/python3.10/site-packages/diffusers/pipelines/pipeline_utils.py", line 571, in register_modules
library = not_compiled_module.__module__.split(".")[0]
AttributeError: 'bool' object has no attribute '__module__'
This is similar to the following error:
huggingface/diffusers#6094
The solution is discussed here:
huggingface/diffusers#5993 (comment)
Are you interested in solving this and making MIGC compatible with the updated diffusers?
Is the image size generated by this method fixed?
Hello, thanks for the excellent work, may I ask when the training code will be released, if it will be released soon, thank you!
can not draw box
A great job, are there any tips on setting up bounding boxes to perform batch inferencing?
Did the validation set of COCO-2014 included in the training stage of your model?
hello,thanks for your excellent work, will you release the pretrained weights of sdxl in the future?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.