deforum-art / deforum-stable-diffusion

Deforum Stable Diffusion

  • Public
  • 73.6K runs
  • A100 (80GB)
  • GitHub
  • License
  • Prediction

    deforum-art/deforum-stable-diffusion:528a34416002e74ade02a83781de899e7101d00ca9008f78e0f4217adc13137c
    ID
    rvhvjhotu5b63iadeui33cyhsi
    Status
    Succeeded
    Source
    Web
    Hardware
    Total duration
    Created
    by @chenxwh

    Input

    fov
    40
    fps
    15
    zoom
    0:(1.04)
    angle
    0:(0)
    width
    512
    border
    replicate
    height
    512
    sampler
    euler_ancestral
    clip_name
    ViT-L/14
    far_plane
    10000
    max_frames
    200
    near_plane
    200
    midas_weight
    0.3
    padding_mode
    border
    rotation_3d_x
    0:(0)
    rotation_3d_y
    0:(0)
    rotation_3d_z
    0:(0)
    sampling_mode
    bicubic
    translation_x
    0:(10*sin(2*3.14*t/10))
    translation_y
    0:(0)
    translation_z
    0:(10)
    animation_mode
    2D
    guidance_scale
    7
    noise_schedule
    0: (0.02)
    sigma_schedule
    0: (1.0)
    amount_schedule
    0: (0.2)
    color_coherence
    Match Frame 0 LAB
    kernel_schedule
    0: (5)
    model_checkpoint
    Protogen_V2.2.ckpt
    animation_prompts
    0: a beautiful apple, trending on Artstation | 50: a beautiful banana, trending on Artstation | 100: a beautiful coconut, trending on Artstation | 150: a beautiful durian, trending on Artstation
    contrast_schedule
    0: (1.0)
    diffusion_cadence
    1
    extract_nth_frame
    1
    resume_timestring
    20220829210106
    strength_schedule
    0: (0.65)
    use_depth_warping
    threshold_schedule
    0: (0.0)
    hybrid_video_motion
    None
    num_inference_steps
    50
    perspective_flip_fv
    0:(53)
    interpolate_x_frames
    4
    perspective_flip_phi
    0:(t%15)
    perspective_flip_gamma
    0:(0)
    perspective_flip_theta
    0:(0)
    hybrid_video_flow_method
    Farneback
    overwrite_extracted_frames
    hybrid_video_comp_mask_type
    None
    hybrid_video_comp_mask_equalize
    None
    hybrid_video_comp_alpha_schedule
    0:(1)
    color_coherence_video_every_N_frames
    1
    hybrid_video_comp_mask_contrast_schedule
    0:(1)
    hybrid_video_use_first_frame_as_init_image
    hybrid_video_comp_mask_blend_alpha_schedule
    0:(0.5)
    hybrid_video_comp_mask_auto_contrast_cutoff_low_schedule
    0:(0)
    hybrid_video_comp_mask_auto_contrast_cutoff_high_schedule
    0:(100)

    Output

    Generated in
  • Prediction

    deforum-art/deforum-stable-diffusion:528a34416002e74ade02a83781de899e7101d00ca9008f78e0f4217adc13137c
    ID
    u46w5dzgtrfg3aeqnvka563l7e
    Status
    Succeeded
    Source
    Web
    Hardware
    Total duration
    Created
    by @chenxwh

    Input

    fov
    40
    fps
    15
    zoom
    0:(1.05), 100:(0.95), 180:(1.05)
    angle
    0:(1), 100:(-3), 200:(1), 300:(-1)
    width
    512
    border
    replicate
    height
    "512"
    sampler
    euler_ancestral
    clip_name
    ViT-L/14
    far_plane
    10000
    max_frames
    400
    near_plane
    200
    midas_weight
    0.3
    padding_mode
    border
    rotation_3d_x
    0:(0)
    rotation_3d_y
    0:(0)
    rotation_3d_z
    0:(0)
    sampling_mode
    bicubic
    translation_x
    0:(10*sin(2*3.14*t/10))
    translation_y
    0:(0)
    translation_z
    0:(10)
    animation_mode
    2D
    guidance_scale
    7
    noise_schedule
    0: (0.02)
    sigma_schedule
    0: (1.0)
    amount_schedule
    0: (0.2)
    color_coherence
    Match Frame 0 LAB
    kernel_schedule
    0: (5)
    model_checkpoint
    Protogen_V2.2.ckpt
    animation_prompts
    0: a beautiful forest by Asher Brown Durand, trending on Artstation | 100: a beautiful portrait of a woman by Artgerm, trending on Artstation | 200: female cyborg assimilated by alien fungus, intricate Three-point lighting portrait, by Ching Yeh and Greg Rutkowski, detailed cyberpunk in the style of GitS 1995 | 300: concept art of a far-future city, key visual, summer day, highly detailed, digital painting, artstation, concept art, sharp focus, in harmony with nature, streamlined, by makoto shinkai and akihiko yoshida and hidari and wlop
    contrast_schedule
    0: (1.0)
    diffusion_cadence
    1
    extract_nth_frame
    1
    strength_schedule
    0: (0.65)
    use_depth_warping
    threshold_schedule
    0: (0.0)
    hybrid_video_motion
    None
    num_inference_steps
    50
    perspective_flip_fv
    0:(53)
    interpolate_x_frames
    4
    perspective_flip_phi
    0:(t%15)
    perspective_flip_gamma
    0:(0)
    perspective_flip_theta
    0:(0)
    hybrid_video_flow_method
    Farneback
    overwrite_extracted_frames
    hybrid_video_comp_mask_type
    None
    hybrid_video_comp_mask_equalize
    None
    hybrid_video_comp_alpha_schedule
    0:(1)
    color_coherence_video_every_N_frames
    1
    hybrid_video_comp_mask_contrast_schedule
    0:(1)
    hybrid_video_use_first_frame_as_init_image
    hybrid_video_comp_mask_blend_alpha_schedule
    0:(0.5)
    hybrid_video_comp_mask_auto_contrast_cutoff_low_schedule
    0:(0)
    hybrid_video_comp_mask_auto_contrast_cutoff_high_schedule
    0:(100)

    Output

    Generated in
  • Prediction

    deforum-art/deforum-stable-diffusion:652b0fed80b8c0845b20de06f877115f56b70b2136d02db95f163eff4b95e35d
    ID
    wnnsa63af5cwpkxesthignc7du
    Status
    Succeeded
    Source
    Web
    Hardware
    Total duration
    Created
    by @chenxwh

    Input

    fov
    140
    fps
    "30"
    zoom
    0:(1.04)
    angle
    0:(0)
    width
    "768"
    border
    wrap
    height
    "384"
    sampler
    euler_ancestral
    clip_name
    ViT-L/14
    far_plane
    10000
    max_frames
    1000
    near_plane
    200
    midas_weight
    0.3
    padding_mode
    border
    rotation_3d_x
    0:(0)
    rotation_3d_y
    0:(-1*sin(2*3.14*t/1200))
    rotation_3d_z
    0:(0)
    sampling_mode
    bicubic
    translation_x
    0:(5*sin(2*3.14*t/1200))
    translation_y
    0:(-0.5), 150:(0.5*sin(2*3.14*t/600))
    translation_z
    0:(-0.5)
    animation_mode
    3D
    guidance_scale
    "9"
    noise_schedule
    0: (0.08*(cos(3.141*t/15)**1000)+0.02)
    sigma_schedule
    0: (1.0)
    amount_schedule
    0: (0.2)
    color_coherence
    Match Frame 0 LAB
    kernel_schedule
    0: (5)
    model_checkpoint
    Protogen_V2.2.ckpt
    animation_prompts
    mdjrny-v4 style a highly detailed matte painting of some attractive young witches doing pagan rites around a bonefire in a forgoten forest at night time, moonlight, magical lighting, by greg rutkowski, dramatic, epic details, close up, realistic, sharp, 4 k resolution, trending on artstation
    contrast_schedule
    0: (1.0)
    diffusion_cadence
    5
    extract_nth_frame
    1
    strength_schedule
    0: (0.65), 5:(-0.15*(cos(3.141*t/50)**100)+0.65)
    use_depth_warping
    threshold_schedule
    0: (0.0)
    hybrid_video_motion
    None
    num_inference_steps
    "100"
    perspective_flip_fv
    0:(53)
    interpolate_x_frames
    4
    perspective_flip_phi
    0:(t%15)
    perspective_flip_gamma
    0:(0)
    perspective_flip_theta
    0:(0)
    hybrid_video_flow_method
    Farneback
    overwrite_extracted_frames
    hybrid_video_comp_mask_type
    None
    hybrid_video_comp_mask_equalize
    None
    hybrid_video_comp_alpha_schedule
    0:(1)
    color_coherence_video_every_N_frames
    1
    hybrid_video_comp_mask_contrast_schedule
    0:(1)
    hybrid_video_use_first_frame_as_init_image
    hybrid_video_comp_mask_blend_alpha_schedule
    0:(0.5)
    hybrid_video_comp_mask_auto_contrast_cutoff_low_schedule
    0:(0)
    hybrid_video_comp_mask_auto_contrast_cutoff_high_schedule
    0:(100)

    Output

    Generated in

Want to make some of these yourself?

Run this model