MACE icon indicating copy to clipboard operation
MACE copied to clipboard

[CVPR 2024] "MACE: Mass Concept Erasure in Diffusion Models" (Official Implementation)

Results 9 MACE issues
Sort by recently updated
recently updated
newest added

Hi, I'm looking to reproduce the performance of the celebrity 1,5,10,100 erasure models in the paper and the ones on OneDrive. I was wondering what the hyperparameters used to train...

Hi, I am looking to replicate the results from the celebrity experiments in the paper using this repo. I'm a bit confused as to what dataset is used to compare...

GIPHY Celebrity Detector Installation Guide, after install you may encounter 'APP_DATA_DIR' is none 'APP_RECOGNITION_WEIGHTS_FILE' is None error. All you have to do is to export the APP_DATA_DIR=celeb-detection-oss/examples/resources APP_RECOGNITION_WEIGHTS_FILE=face_recognition/best_model_states.pkl

Hello,Your work is great.But the output effect of SD1.4 version is not ideal, the quality is relatively poor, is there any Mass Concept Erasure on SDXL version?

Thank you for this solid and interesting work. However, I found that the clip_score evaluation script of the 100_arts style is not included in this repository. The description of the...

Hello, I have installed most dependencies following your instructions. But in the last step installing recognize-anything by the command "pip install -e ./recognize-anything/", it meets an error: ![image](https://github.com/Shilin-LU/MACE/assets/62345916/a19351d8-5b48-4a90-9777-eef221f44152) Could you...

Hi, I am facing some problem about the installation. I do following command git clone https://github.com/Shilin-LU/MACE.git conda create -n mace python=3.10 conda activate mace conda install pytorch==2.0.1 torchvision==0.15.2 pytorch-cuda=11.7 -c...

Hi, thanks for this wonderful work! I doubt the process of NSFW prompts and masked images. Because these images do not contain explicit region boundaries. For example, How to mask...

Dear Shilin-LU I am very much looking forward to this work. I would like to inquire about when it will be available on arXiv? thanks