Stability-AI / generative-models

Generative Models by Stability AI

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

run sv3d_u success but output mp4 can't play

hengshan123 opened this issue · comments

hi, I run the script success:
python scripts/sampling/simple_video_sample.py --input_path assets/000.jpg --version sv3d_u

Initialized embedder #0: FrozenOpenCLIPImagePredictionEmbedder with 683800065 params. Trainable: False
Initialized embedder #1: VideoPredictionEmbedderWithEncoder with 83653863 params. Trainable: False
Initialized embedder #2: ConcatTimestepEmbedderND with 0 params. Trainable: False
Restored from checkpoints/sv3d_u.safetensors with 0 missing and 0 unexpected keys
############################## Sampling setting ##############################
Sampler: EulerEDMSampler
Discretization: EDMDiscretization
Guider: TrianglePredictionGuider
Sampling with EulerEDMSampler for 51 steps: 0%| | 0/51 [00:00<?, ?it/s]/home/j00668593/anaconda3/envs/py310/lib/python3.10/site-packages/torch/utils/checkpoint.py:31: UserWarning: None of the inputs have requires_grad=True. Gradients will be None
warnings.warn("None of the inputs have requires_grad=True. Gradients will be None")
Sampling with EulerEDMSampler for 51 steps: 98%|██████████████████████████████████████████████████████████████████████████████████████████▏ | 50/51 [00:59<00:01, 1.19s/it]

And i get the output in outputs/simple_video_sample/sv3d_u.
But the 000000.mp4 can't be played.

Same here for sv3d_p

same here for sv3d_u

I check the origin result vid, this is valid. the problem maybe in imageio
image

confirmed the issue is from imageio. Switch to cv2 solved the problem. fix: #327

confirmed the issue is from imageio. Switch to cv2 solved the problem. fix: #327

Thanks for your advise, the issue is fixed !

after install imageio-ffmpeg the problem is solved

same question