search

magic-research / magic-animate

[CVPR 2024] MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model
https://showlab.github.io/magicanimate/
BSD 3-Clause "New" or "Revised" License
10.4k stars 1.07k forks source link

how to save save_individual_videos #52

Open magicwang1111 opened 10 months ago

magicwang1111 commented 10 months ago

i only want to save the final video,i have changed animation.yaml。 `pretrained_model_path: "pretrained_models/stable-diffusion-v1-5" pretrained_vae_path: "pretrained_models/sd-vae-ft-mse" pretrained_controlnet_path: "pretrained_models/MagicAnimate/densepose_controlnet" pretrained_appearance_encoder_path: "pretrained_models/MagicAnimate/appearance_encoder" pretrained_unet_path: ""

motion_module: "pretrained_models/MagicAnimate/temporal_attention/temporal_attention.ckpt"

savename: null

fusion_blocks: "midup"

seed: [1] steps: 25 guidance_scale: 7.5

source_image:

inference_config: "configs/inference/inference.yaml" size: 512 L: 16 S: 1 I: 0 clip: 0 offset: 0 max_length: null video_type: "condition" invert_video: false save_individual_videos: true`

but save_individual_videos not work

FurkanGozukara commented 10 months ago

not only save_individual_videos but also size doesn't work

natlamir commented 10 months ago

When run with the inference command, It is saving the final video for me under samples/TIMESTAMPFOLDER/videos

SoftologyPro commented 10 months ago

Please save the individual video when using the gradio demo too. Save into the same output directory as the combined output video.

FurkanGozukara commented 10 months ago

Please save the individual video when using the gradio demo too. Save into the same output directory as the combined output video.

check here : https://github.com/magic-research/magic-animate/issues/44

i added this option

i plan to add face restore hopefully today

image

SoftologyPro commented 10 months ago

Will this be paywalled forever, or be merged into the git soon?