mirror of
https://github.com/storytold/storyteller-ml.git
synced 2026-10-28 09:08:54 +00:00
Brandon Thomas
046ab789c0
https://lucaria-academy.github.io/Animate-X/ https://github.com/antgroup/animate-x @ fdc80909f911d8e487cb4e6847f2e4c6501b62af
111 lines
3.1 KiB
YAML
111 lines
3.1 KiB
YAML
|
|
|
|
# manual setting
|
|
max_frames: 32
|
|
resolution: [512, 768]
|
|
round: 5
|
|
ddim_timesteps: 30
|
|
seed: 13
|
|
test_list_path: [
|
|
[2, "data/images/1.jpg", "data/saved_pose/dance_1","data/saved_frames/dance_1","data/saved_pkl/dance_1.pkl", 14], # seed 14
|
|
[2, "data/images/4.png", "data/saved_pose/dance_1","data/saved_frames/dance_1","data/saved_pkl/dance_1.pkl", 14], # seed 14
|
|
[2, "data/images/3.jpeg", "data/saved_pose/dance_2","data/saved_frames/dance_2","data/saved_pkl/dance_2.pkl", 13], # seed 13
|
|
[2, "data/images/10.jpeg", "data/saved_pose/dance_2","data/saved_frames/dance_2","data/saved_pkl/dance_2.pkl", 14], #
|
|
[2, "data/images/zhu.png", "data/saved_pose/dance_3","data/saved_frames/dance_3","data/saved_pkl/dance_3.pkl", 14],
|
|
[2, "data/images/5.jpeg", "data/saved_pose/dance_3","data/saved_frames/dance_3","data/saved_pkl/dance_3.pkl", 14],
|
|
[1, "data/images/2.jpeg", "data/saved_pose/dance_3_right","data/saved_frames/dance_3_right","data/saved_pkl/dance_3_right.pkl", 13], # seed 13
|
|
[4, "data/images/7.jpeg", "data/saved_pose/ubc","data/saved_frames/ubc","data/saved_pkl/ubc.pkl", 14],
|
|
]
|
|
|
|
|
|
log_dir: 'results' # save dir
|
|
|
|
test_model: checkpoints/animate-x_ckpt.pth
|
|
partial_keys: [
|
|
['image','local_image', 'dwpose','pose_embeddings'],
|
|
]
|
|
|
|
|
|
|
|
|
|
# default settings
|
|
TASK_TYPE: inference_animate_x_entrance
|
|
use_fp16: True
|
|
guide_scale: 2.5
|
|
vit_resolution: [224, 224]
|
|
use_fp16: True
|
|
batch_size: 1
|
|
latent_random_ref: True
|
|
chunk_size: 2
|
|
decoder_bs: 2
|
|
scale: 8
|
|
use_fps_condition: False
|
|
|
|
embedder: {
|
|
'type': 'FrozenOpenCLIPTextVisualEmbedder',
|
|
'layer': 'penultimate',
|
|
'pretrained': 'checkpoints/open_clip_pytorch_model.bin'
|
|
}
|
|
|
|
auto_encoder: {
|
|
'type': 'AutoencoderKL',
|
|
'ddconfig': {
|
|
'double_z': True,
|
|
'z_channels': 4,
|
|
'resolution': 256,
|
|
'in_channels': 3,
|
|
'out_ch': 3,
|
|
'ch': 128,
|
|
'ch_mult': [1, 2, 4, 4],
|
|
'num_res_blocks': 2,
|
|
'attn_resolutions': [],
|
|
'dropout': 0.0,
|
|
'video_kernel_size': [3, 1, 1]
|
|
},
|
|
'embed_dim': 4,
|
|
'pretrained': 'checkpoints/v2-1_512-ema-pruned.ckpt'
|
|
}
|
|
|
|
|
|
UNet: {
|
|
'type': 'UNetSD_Animate_X',
|
|
'config': None,
|
|
'in_dim': 4,
|
|
'num': 0,
|
|
'no_hand': True,
|
|
'dim': 320,
|
|
'y_dim': 1024,
|
|
'context_dim': 1024,
|
|
'out_dim': 4,
|
|
'dim_mult': [1, 2, 4, 4],
|
|
'num_heads': 8,
|
|
'head_dim': 64,
|
|
'num_res_blocks': 2,
|
|
'dropout': 0.1,
|
|
'temporal_attention': True,
|
|
'num_tokens': 4,
|
|
'temporal_attn_times': 1,
|
|
'use_checkpoint': True,
|
|
'use_fps_condition': False,
|
|
'use_sim_mask': False
|
|
}
|
|
video_compositions: ['image', 'local_image', 'dwpose', 'randomref', 'randomref_pose', 'pose_embedding']
|
|
Diffusion: {
|
|
'type': 'DiffusionDDIM',
|
|
'schedule': 'linear_sd',
|
|
'schedule_param': {
|
|
'num_timesteps': 1000,
|
|
"init_beta": 0.00085,
|
|
"last_beta": 0.0120,
|
|
'zero_terminal_snr': True,
|
|
},
|
|
'mean_type': 'v',
|
|
'loss_type': 'mse',
|
|
'var_type': 'fixed_small',
|
|
'rescale_timesteps': False,
|
|
'noise_strength': 0.1
|
|
}
|
|
use_DiffusionDPM: False
|
|
CPU_CLIP_VAE: True
|
|
|