demo/configs/i2v_config.yaml

prompts:
- - lightning, lighthouse
  # - sun rising, lighthouse
  # - fireworks, lighthouse

n_prompt:
  - 'wrong white balance, dark, sketches,worst quality,low quality, deformed, distorted, disfigured, bad eyes, wrong lips,weird mouth, bad teeth, mutated hands and fingers, bad anatomy,wrong anatomy, amputation, extra limb, missing limb, floating,limbs, disconnected limbs, mutation, ugly, disgusting, bad_pictures, negative_hand-neg'

generate:
  model_path: "checkpoints/i2v/unet/pia.ckpt"
  use_image: true
  use_video: false
  sample_width: 512
  sample_height: 512
  video_length: 16
  use_lora: false
  use_db: true
  global_seed: 5658137986800322011
  lora_path: ""
  db_path: "checkpoints/i2v/dreambooth/rcnzCartoon3d_v10.safetensors"
  lora_alpha: 0.8

validation_data:
  # mask_sim_range: [0, 1]
  mask_sim_range: [0]
  cond_frame: 0
  num_inference_steps: 25

  img_mask: ''
  input_name: 'lighthouse'
  validation_input_path: 'img'
  save_path: 'result'

noise_scheduler_kwargs:
  num_train_timesteps: 1000
  beta_start:          0.00085
  beta_end:            0.012
  beta_schedule:       "linear"
  steps_offset:        1
  clip_sample:         false

pretrained_model_path: "checkpoints/diffusion_body/stable-diffusion-v1-5"
unet_additional_kwargs:
  use_motion_module              : true
  motion_module_resolutions      : [ 1,2,4,8 ]
  unet_use_cross_frame_attention : false
  unet_use_temporal_attention    : false

  motion_module_type: Vanilla
  motion_module_kwargs:
    num_attention_heads                : 8
    num_transformer_block              : 1
    attention_block_types              : [ "Temporal_Self", "Temporal_Self" ]
    temporal_position_encoding         : true
    temporal_position_encoding_max_len : 32
    temporal_attention_dim_div         : 1
    zero_initialize                    : true