Bug: 3D mode broken in Deforum?

When switching to Deforum's animation mode to 3D, I encountered a strange error

Steps to reproduce the problem:

Go to Deforum tab Press Keyframes Change animation mode to 3D Change border to wrap Press Generate

My CLA: @echo off

set PYTHON=C:\Python310\python.exe set GIT= set VENV_DIR= set COMMANDLINE_ARGS= --xformers --disable-safe-unpickle --allow-code --autolaunch --opt-channelslast --skip-torch-cuda-test --theme dark set CUDA_VISIBLE_DEVICES=0 set SAFETENSORS_FAST_GPU=1

git pull

call webui.bat

Commit hash: https://github.com/AUTOMATIC1111/stable-diffusion-webui/commit/151233399c4b79934bdbb7c12a97eeb6499572fb

LOG:

Deforum script for 2D, pseudo-2D and 3D animations v0.5-webui-beta Additional models path: G:\StableDiffusion\stable-diffusion-webui\models/Deforum Saving animation frames to G:\StableDiffusion\stable-diffusion-webui\outputs/img2img-images\Deforum Using cache found in C:\Users\rethink/.cache\torch\hub\rwightman_gen-efficientnet-pytorch_master Loading base model ()...Done. Removing last two layers (global_pool & classifier). Building Encoder-Decoder model..Done. Error completing request Arguments: (False, '', '3D', 120, 'wrap', '0:(0)', '0:(1.02+0.02sin(23.14t/20))', '0:(0)', '0:(0)', '0:(10)', '0:(0)', '0:(0)', '0:(0)', False, '0:(0)', '0:(t%15)', '0:(0)', '0:(53)', '0: (0.08)', '0: (0.6)', '0: (1.0)', '0: (7)', False, '0: (25)', '0: (40)', '0: (200)', '0: (10000)', '0: (t%4294967293)', False, '0: ("Euler a")', '0: (5)', '0: (1.0)', '0: (0.2)', '0: (0.0)', True, 'Match Frame 0 LAB', 1.0, 1.0, 'perlin', 8.0, 8.0, 4, 0.5, True, 0.3, 200.0, 10000.0, 40.0, 'border', 'bicubic', False, '/content/video_in.mp4', 1, False, False, '/content/video_in.mp4', False, 4, False, '20220829210106', '[\n "a beautiful forest by Asher Brown Durand, trending on Artstation",\n "a beautiful portrait of a woman by Artgerm, trending on Artstation"\n]\n', '{\n "0": "(scenic countryside:1.0), (cherry:`where(cos(6.28t/10)>0, 1.8cos(6.28t/10), 0.001)), (strawberry:where(cos(6.28t/10)<0, -1.8cos(6.28t/10), 0.001)), snow, detailed painting by greg rutkowski --neg (cherry:where(cos(6.28t/10)<0, -1.8cos(6.28t/10), 0.001)), (strawberry:where(cos(6.28t/10)>0, 1.8cos(6.28t/10), 0.001))",\n "60": "a beautiful (((banana))), trending on Artstation",\n "80": "a beautiful coconut --neg photo, realistic",\n "100": "a beautiful durian, trending on Artstation"\n}\n', 512, 512, False, False, False, 0, 0, -1, 'Euler a', False, -1, 0, 0, 0, 21, 0.0, 1, False, 1, True, True, False, False, False, False, 'Deforum', '{timestring}_{index}_{prompt}.png', 'iter', 1, False, False, True, 0, 'https://user-images.githubusercontent.com/14872007/195867706-d067cdc6-28cd-450b-a61e-55e25bc67010.png', False, False, False, True, 'https://www.filterforge.com/wiki/images/archive/b/b7/20080927223728%21Polygonal_gradient_thumb.jpg', 1.0, 1.0, 5.0, 1, True, 4, 'reroll', False, 12.0, 'PIL gif', 'ffmpeg', False, 'snowfall.mp3', False, False, 200.0, 'x0_pred', '/content/drive/MyDrive/AI/StableDiffusion/2022-09/20220903000939_%05d.png', '/content/drive/MyDrive/AI/StableDiffusion/content/drive/MyDrive/AI/StableDiffusion/2022-09/kabachuha/2022-09/20220903000939.mp4', False, '', True, False, False, 'None', 'Farneback', False, 'None', False, 'None', False, False, '0:(1)', '0:(0.5)', '0:(1)', '0:(100)', '0:(0)', 'Deforum v0.5-webui-beta', 'Made by <a href="https://deforum.github.io">deforum.github.io</a>, port for AUTOMATIC1111\'s webui maintained by <a href="https://github.com/kabachuha">kabachuha</a>', '<ul style="list-style-type:circle; margin-left:1em">\n <li>The code for this extension: <a style="color:blue" href="https://github.com/deforum-art/deforum-for-automatic1111-webui">Fork of deforum for auto1111\'s webui</a>.</li>\n <li>Join the <a style="color:blue" href="https://discord.gg/deforum">official Deforum Discord</a> to share your creations and suggestions.</li>\n <li>For general usage, see the <a style="color:blue" href="https://docs.google.com/document/d/1pEobUknMFMkn8F5TMsv8qRzamXX_75BShMMXV8IFslI/edit">User guide for Deforum v0.5</a>.</li>\n <li>For advanced animations, see the <a style="color:blue" href="https://docs.google.com/document/d/1pfW1PwbDIuW0cv-dnuyYj1UzPqe23BlSLTJsqazffXM/edit?usp=sharing">Math keyframing explanation</a>.</li>\n <li>Alternatively, use <a style="color:blue" href="https://sd-parseq.web.app/deforum">sd-parseq</a> as a UI to define your animation schedules (see the Parseq section in the Keyframes tab).</li>\n <li><a style="color:blue" href="https://www.framesync.xyz/">framesync.xyz</a> is also a good option, it makes compact math formulae for Deforum keyframes by selecting various waveforms.</li>\n <li>The other site allows for making keyframes using <a style="color:blue" href="https://www.chigozie.co.uk/keyframe-string-generator/">interactive splines and Bezier curves</a> (select Disco output format).</li>\n <li>After the 2022-12-30 update, the default noise type is <a style="color:blue" href="https://en.wikipedia.org/wiki/Perlin_noise">Perlin noise</a> due to its great frame coherence improvements. If you want to use the old noise and replicate the previous settings, set the type to "uniform" in the Keyframes tab.</li>\n </ul>\n <italic>If you liked this extension, please <a style="color:blue" href="https://github.com/deforum-art/deforum-for-automatic1111-webui">give it a star on GitHub</a>!</italic> 😊', '', '', '', '', 'Import settings from file', 'Animation settings', 'Motion parameters:', '2D and 3D settings', 'Prespective flip — Low VRAM pseudo-3D mode:', 'Coherence:', '3D Depth Warping:', 'Video Input:', 'Interpolation (turned off atm)', 'Resume animation:', 'Prompts', 'animation_mode: None` batches on list of prompts. (Batch mode disabled atm, only animation_prompts are working)

', '

Important change from vanilla Deforum!

', '
This script uses the built-in webui weighting settings.
', '
So if you want to use math functions as prompt weights,
', '
keep the values above zero in both parts
', '
Negative prompt part can be specified with --neg
', '
Run settings
', '
Sampling settings
', '
3D Fov settings:
', '', '', '', '
Batch settings
', '
Init settings
', '
3D settings
', '
Steps scheduling:
', '
Video output settings
', '
To enable seed schedule select seed behavior — \'schedule\'
', '\n Use an <a style=\'color:blue;\' target=\'_blank\' href=\'https://sd-parseq.web.app/deforum\'>sd-parseq manifest for your animation (leave blank to ignore).
\n
\n Note that parseq overrides:\n
\n
Run: seed, subseed, subseed strength.
\n
Keyframes: generation settings (noise, strength, contrast, scale).
\n
Keyframes: motion parameters for 2D and 3D (angle, zoom, translation, rotation, perspective flip).
\n
\n
\n <p">\n Parseq does not override:\n
\n
Run: Sampler, W, H, Restore faces, tiling, highres fix, resize seed.
\n
Keyframes: animation settings (animation mode, max_frames, border)
\n
Keyframes: coherence (color coherence & diffusion cadence)
\n
Keyframes: depth warping
\n
Video output settings: all settings (including fps and max frames)
\n
\n
\n ', '
Anti-blur settings
', '
Perlin noise params, if selected.
', '
Hybrid Video Compositing in 2D/3D Modeby reallybigname
', '
Composite video with previous frame init image in 2D or 3D animation_mode (not for Video Input mode)
Uses your Init settings for video_init_path, extract_nth_frame, overwrite_extracted_frames
In Keyframes tab, you can also set color_coherence = \'Video Input\'
color_coherence_video_every_N_frames lets you only match every N frames
Color coherence may be used with hybrid composite off, to just use video color.
Hybrid motion may be used with hybrid composite off, to just use video motion.
Hybrid Video Schedules
The alpha schedule controls overall alpha for video mix, whether using a composite mask or not.
The hybrid_comp_mask_blend_alpha_schedule only affects the \'Blend\' hybrid_comp_mask_type.
Mask contrast schedule is from 0-255. Normal is 1. Affects all masks.
Autocontrast low/high cutoff schedules 0-100. Low 0 High 100 is full range.
(hybrid_comp_mask_auto_contrast must be enabled)
See main Deforum document for more details.', '
Hybrid Video Schedules
') {} Traceback (most recent call last): File "G:\StableDiffusion\stable-diffusion-webui\modules\call_queue.py", line 45, in f res = list(func(args, kwargs)) File "G:\StableDiffusion\stable-diffusion-webui\modules\call_queue.py", line 28, in f res = func(*args, *kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\deforum-for-automatic1111-webui\scripts\deforum.py", line 255, in run_deforum processed = DeforumScript.run(None, p, override_settings_with_file, custom_settings_file, animation_mode, max_frames, border, angle, zoom, translation_x, translation_y, translation_z, rotation_3d_x, rotation_3d_y, rotation_3d_z, flip_2d_perspective, perspective_flip_theta, perspective_flip_phi, perspective_flip_gamma, perspective_flip_fv, noise_schedule, strength_schedule, contrast_schedule, cfg_scale_schedule, enable_steps_scheduling, steps_schedule, fov_schedule, near_schedule, far_schedule, seed_schedule, enable_sampler_scheduling, sampler_schedule, kernel_schedule, sigma_schedule, amount_schedule, threshold_schedule, histogram_matching, color_coherence, color_coherence_video_every_N_frames, diffusion_cadence, noise_type, perlin_w, perlin_h, perlin_octaves, perlin_persistence, use_depth_warping, midas_weight, near_plane, far_plane, fov, padding_mode, sampling_mode, save_depth_maps, video_init_path, extract_nth_frame, overwrite_extracted_frames, use_mask_video, video_mask_path, interpolate_key_frames, interpolate_x_frames, resume_from_timestring, resume_timestring, prompts, animation_prompts, W, H, restore_faces, tiling, enable_hr, firstphase_width, firstphase_height, seed, sampler, seed_enable_extras, subseed, subseed_strength, seed_resize_from_w, seed_resize_from_h, steps, ddim_eta, n_batch, make_grid, grid_rows, save_settings, save_samples, display_samples, save_sample_per_step, show_sample_per_step, override_these_with_webui, batch_name, filename_format, seed_behavior, seed_iter_N, use_init, from_img2img_instead_of_link, strength_0_no_init, strength, init_image, use_mask, use_alpha_as_mask, invert_mask, overlay_mask, mask_file, mask_contrast_adjust, mask_brightness_adjust, mask_overlay_blur, fill, full_res_mask, full_res_mask_padding, reroll_blank_frames, skip_video_for_run_all, fps, output_format, ffmpeg_location, add_soundtrack, soundtrack_path, use_manual_settings, render_steps, max_video_frames, path_name_modifier, image_path, mp4_path, store_frames_in_ram, parseq_manifest, parseq_use_deltas, hybrid_generate_inputframes, hybrid_use_first_frame_as_init_image, hybrid_motion, hybrid_flow_method, hybrid_composite, hybrid_comp_mask_type, hybrid_comp_mask_inverse, hybrid_comp_mask_equalize, hybrid_comp_mask_auto_contrast, hybrid_comp_save_extra_frames, hybrid_comp_alpha_schedule, hybrid_comp_mask_blend_alpha_schedule, hybrid_comp_mask_contrast_schedule, hybrid_comp_mask_auto_contrast_cutoff_high_schedule, hybrid_comp_mask_auto_contrast_cutoff_low_schedule, i1, i2, i3, i4, i5, i6, i7, i8, i9, i10, i11, i12, i13, i14, i15, i16, i17, i18, i19, i20, i21, i22, i23, i24, i25, i26, i27, i28, i29, i30, i31, i32, i33, i34, i35, i36, i37, i38, i39, i40, i41, i42) File "G:\StableDiffusion\stable-diffusion-webui\extensions\deforum-for-automatic1111-webui\scripts\deforum.py", line 93, in run render_animation(args, anim_args, video_args, parseq_args, root.animation_prompts, root) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum-for-automatic1111-webui/scripts\deforum_helpers\render.py", line 92, in render_animation depth_model.load_adabins(root.models_path) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum-for-automatic1111-webui/scripts\deforum_helpers\depth.py", line 38, in load_adabins self.adabins_helper = InferenceHelper(models_path=models_path, dataset='nyu', device=self.device) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum-for-automatic1111-webui/scripts/deforumhelpers/src\infer.py", line 85, in init model, , _ = model_io.load_checkpoint(pretrained_path, model) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum-for-automatic1111-webui/scripts/deforum_helpers/src\model_io.py", line 37, in load_checkpoint ckpt = torch.load(fpath, map_location='cpu') File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_smartprocess\reallysafe.py", line 117, in load return load_with_extra(filename, args, kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_smartprocess\reallysafe.py", line 164, in load_with_extra return unsafe_torch_load(filename, *args, kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_dreambooth_extension\reallysafe.py", line 117, in load return load_with_extra(filename, *args, *kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_dreambooth_extension\reallysafe.py", line 164, in load_with_extra return unsafe_torch_load(filename, args, kwargs) File "G:\StableDiffusion\stable-diffusion-webui\modules\safe.py", line 106, in load return load_with_extra(filename, extra_handler=global_extra_handler, *args, *kwargs) File "G:\StableDiffusion\stable-diffusion-webui\modules\safe.py", line 151, in load_with_extra return unsafe_torch_load(filename, args, **kwargs) File "G:\StableDiffusion\stable-diffusion-webui\venv\lib\site-packages\torch\serialization.py", line 705, in load with _open_zipfile_reader(opened_file) as opened_zipfile: File "G:\StableDiffusion\stable-diffusion-webui\venv\lib\site-packages\torch\serialization.py", line 242, in init super(_open_zipfile_reader, self).init(torch._C.PyTorchFileReader(name_or_buffer)) RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

Sounds like your depth models hadn't loaded correctly. Try redownloading it manually using one of the official links (listed in the Readme)

Redownloaded and this is the error: The error looks the same...

Deforum script for 2D, pseudo-2D and 3D animations v0.5-webui-beta Additional models path: G:\StableDiffusion\stable-diffusion-webui\models/Deforum Saving animation frames to G:\StableDiffusion\stable-diffusion-webui\outputs/img2img-images\Deforum_TEST_2023 Using cache found in C:\Users\rethink/.cache\torch\hub\rwightman_gen-efficientnet-pytorch_master Loading base model ()...Done. Removing last two layers (global_pool & classifier). Building Encoder-Decoder model..Done. Error completing request Arguments: (False, '', '3D', 100, 'wrap', '0:(0)', '0:(1.02+0.02sin(23.14t/20))', '0:(0)', '0:(0)', '0:(10)', '0:(0)', '0:(0)', '0:(0)', False, '0:(0)', '0:(t%15)', '0:(0)', '0:(53)', '0: (0.08)', '0: (0.6)', '0: (1.0)', '0: (7)', False, '0: (25)', '0: (40)', '0: (200)', '0: (10000)', '0: (t%4294967293)', False, '0: ("Euler a")', '0: (5)', '0: (1.0)', '0: (0.1)', '0: (0.0)', True, 'Match Frame 0 LAB', 1.0, 1.0, 'perlin', 8.0, 8.0, 4, 0.5, True, 0.3, 200.0, 10000.0, 40.0, 'border', 'bicubic', False, '/content/video_in.mp4', 1, False, False, '/content/video_in.mp4', False, 4, False, '20220829210106', '[\n "a beautiful forest by Asher Brown Durand, trending on Artstation",\n "a beautiful portrait of a woman by Artgerm, trending on Artstation"\n]\n', '{\n "0": "(scenic countryside:1.0), (cherry:`where(cos(6.28t/10)>0, 1.8cos(6.28t/10), 0.001)), (strawberry:where(cos(6.28t/10)<0, -1.8cos(6.28t/10), 0.001)), snow, detailed painting by greg rutkowski --neg (cherry:where(cos(6.28t/10)<0, -1.8cos(6.28t/10), 0.001)), (strawberry:where(cos(6.28t/10)>0, 1.8cos(6.28t/10), 0.001))",\n "60": "a beautiful (((banana))), trending on Artstation",\n "80": "a beautiful coconut --neg photo, realistic",\n "100": "a beautiful durian, trending on Artstation"\n}\n', 512, 512, False, False, False, 0, 0, -1, 'Euler a', False, -1, 0, 0, 0, 21, 0.0, 1, False, 1, True, True, False, False, False, False, 'Deforum_TEST_2023', '{timestring}_{index}_{prompt}.png', 'iter', 1, False, False, True, 0, 'https://user-images.githubusercontent.com/14872007/195867706-d067cdc6-28cd-450b-a61e-55e25bc67010.png', False, False, False, True, 'https://www.filterforge.com/wiki/images/archive/b/b7/20080927223728%21Polygonal_gradient_thumb.jpg', 1.0, 1.0, 5.0, 1, True, 4, 'reroll', True, 12.0, 'PIL gif', 'ffmpeg', False, 'snowfall.mp3', False, False, 200.0, 'x0_pred', '/content/drive/MyDrive/AI/StableDiffusion/2022-09/20220903000939_%05d.png', '/content/drive/MyDrive/AI/StableDiffusion/content/drive/MyDrive/AI/StableDiffusion/2022-09/kabachuha/2022-09/20220903000939.mp4', False, '', True, False, False, 'None', 'Farneback', False, 'None', False, 'None', False, False, '0:(1)', '0:(0.5)', '0:(1)', '0:(100)', '0:(0)', 'Deforum v0.5-webui-beta', 'Made by <a href="https://deforum.github.io">deforum.github.io</a>, port for AUTOMATIC1111\'s webui maintained by <a href="https://github.com/kabachuha">kabachuha</a>', '<ul style="list-style-type:circle; margin-left:1em">\n <li>The code for this extension: <a style="color:blue" href="https://github.com/deforum-art/deforum-for-automatic1111-webui">Fork of deforum for auto1111\'s webui</a>.</li>\n <li>Join the <a style="color:blue" href="https://discord.gg/deforum">official Deforum Discord</a> to share your creations and suggestions.</li>\n <li>For general usage, see the <a style="color:blue" href="https://docs.google.com/document/d/1pEobUknMFMkn8F5TMsv8qRzamXX_75BShMMXV8IFslI/edit">User guide for Deforum v0.5</a>.</li>\n <li>For advanced animations, see the <a style="color:blue" href="https://docs.google.com/document/d/1pfW1PwbDIuW0cv-dnuyYj1UzPqe23BlSLTJsqazffXM/edit?usp=sharing">Math keyframing explanation</a>.</li>\n <li>Alternatively, use <a style="color:blue" href="https://sd-parseq.web.app/deforum">sd-parseq</a> as a UI to define your animation schedules (see the Parseq section in the Keyframes tab).</li>\n <li><a style="color:blue" href="https://www.framesync.xyz/">framesync.xyz</a> is also a good option, it makes compact math formulae for Deforum keyframes by selecting various waveforms.</li>\n <li>The other site allows for making keyframes using <a style="color:blue" href="https://www.chigozie.co.uk/keyframe-string-generator/">interactive splines and Bezier curves</a> (select Disco output format).</li>\n <li>After the 2022-12-30 update, the default noise type is <a style="color:blue" href="https://en.wikipedia.org/wiki/Perlin_noise">Perlin noise</a> due to its great frame coherence improvements. If you want to use the old noise and replicate the previous settings, set the type to "uniform" in the Keyframes tab.</li>\n </ul>\n <italic>If you liked this extension, please <a style="color:blue" href="https://github.com/deforum-art/deforum-for-automatic1111-webui">give it a star on GitHub</a>!</italic> 😊', '', '', '', '', 'Import settings from file', 'Animation settings', 'Motion parameters:', '2D and 3D settings', 'Prespective flip — Low VRAM pseudo-3D mode:', 'Coherence:', '3D Depth Warping:', 'Video Input:', 'Interpolation (turned off atm)', 'Resume animation:', 'Prompts', 'animation_mode: None` batches on list of prompts. (Batch mode disabled atm, only animation_prompts are working)

', '

Important change from vanilla Deforum!

', '
This script uses the built-in webui weighting settings.
', '
So if you want to use math functions as prompt weights,
', '
keep the values above zero in both parts
', '
Negative prompt part can be specified with --neg
', '
Run settings
', '
Sampling settings
', '
3D Fov settings:
', '', '', '', '
Batch settings
', '
Init settings
', '
3D settings
', '
Steps scheduling:
', '
Video output settings
', '
To enable seed schedule select seed behavior — \'schedule\'
', '\n Use an <a style=\'color:blue;\' target=\'_blank\' href=\'https://sd-parseq.web.app/deforum\'>sd-parseq manifest for your animation (leave blank to ignore).
\n
\n Note that parseq overrides:\n
\n
Run: seed, subseed, subseed strength.
\n
Keyframes: generation settings (noise, strength, contrast, scale).
\n
Keyframes: motion parameters for 2D and 3D (angle, zoom, translation, rotation, perspective flip).
\n
\n
\n <p">\n Parseq does not override:\n
\n
Run: Sampler, W, H, Restore faces, tiling, highres fix, resize seed.
\n
Keyframes: animation settings (animation mode, max_frames, border)
\n
Keyframes: coherence (color coherence & diffusion cadence)
\n
Keyframes: depth warping
\n
Video output settings: all settings (including fps and max frames)
\n
\n
\n ', '
Anti-blur settings
', '
Perlin noise params, if selected.
', '
Hybrid Video Compositing in 2D/3D Modeby reallybigname
', '
Composite video with previous frame init image in 2D or 3D animation_mode (not for Video Input mode)
Uses your Init settings for video_init_path, extract_nth_frame, overwrite_extracted_frames
In Keyframes tab, you can also set color_coherence = \'Video Input\'
color_coherence_video_every_N_frames lets you only match every N frames
Color coherence may be used with hybrid composite off, to just use video color.
Hybrid motion may be used with hybrid composite off, to just use video motion.
Hybrid Video Schedules
The alpha schedule controls overall alpha for video mix, whether using a composite mask or not.
The hybrid_comp_mask_blend_alpha_schedule only affects the \'Blend\' hybrid_comp_mask_type.
Mask contrast schedule is from 0-255. Normal is 1. Affects all masks.
Autocontrast low/high cutoff schedules 0-100. Low 0 High 100 is full range.
(hybrid_comp_mask_auto_contrast must be enabled)
See main Deforum document for more details.', '
Hybrid Video Schedules
') {} Traceback (most recent call last): File "G:\StableDiffusion\stable-diffusion-webui\modules\call_queue.py", line 45, in f res = list(func(args, kwargs)) File "G:\StableDiffusion\stable-diffusion-webui\modules\call_queue.py", line 28, in f res = func(*args, *kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\deforum\scripts\deforum.py", line 255, in run_deforum processed = DeforumScript.run(None, p, override_settings_with_file, custom_settings_file, animation_mode, max_frames, border, angle, zoom, translation_x, translation_y, translation_z, rotation_3d_x, rotation_3d_y, rotation_3d_z, flip_2d_perspective, perspective_flip_theta, perspective_flip_phi, perspective_flip_gamma, perspective_flip_fv, noise_schedule, strength_schedule, contrast_schedule, cfg_scale_schedule, enable_steps_scheduling, steps_schedule, fov_schedule, near_schedule, far_schedule, seed_schedule, enable_sampler_scheduling, sampler_schedule, kernel_schedule, sigma_schedule, amount_schedule, threshold_schedule, histogram_matching, color_coherence, color_coherence_video_every_N_frames, diffusion_cadence, noise_type, perlin_w, perlin_h, perlin_octaves, perlin_persistence, use_depth_warping, midas_weight, near_plane, far_plane, fov, padding_mode, sampling_mode, save_depth_maps, video_init_path, extract_nth_frame, overwrite_extracted_frames, use_mask_video, video_mask_path, interpolate_key_frames, interpolate_x_frames, resume_from_timestring, resume_timestring, prompts, animation_prompts, W, H, restore_faces, tiling, enable_hr, firstphase_width, firstphase_height, seed, sampler, seed_enable_extras, subseed, subseed_strength, seed_resize_from_w, seed_resize_from_h, steps, ddim_eta, n_batch, make_grid, grid_rows, save_settings, save_samples, display_samples, save_sample_per_step, show_sample_per_step, override_these_with_webui, batch_name, filename_format, seed_behavior, seed_iter_N, use_init, from_img2img_instead_of_link, strength_0_no_init, strength, init_image, use_mask, use_alpha_as_mask, invert_mask, overlay_mask, mask_file, mask_contrast_adjust, mask_brightness_adjust, mask_overlay_blur, fill, full_res_mask, full_res_mask_padding, reroll_blank_frames, skip_video_for_run_all, fps, output_format, ffmpeg_location, add_soundtrack, soundtrack_path, use_manual_settings, render_steps, max_video_frames, path_name_modifier, image_path, mp4_path, store_frames_in_ram, parseq_manifest, parseq_use_deltas, hybrid_generate_inputframes, hybrid_use_first_frame_as_init_image, hybrid_motion, hybrid_flow_method, hybrid_composite, hybrid_comp_mask_type, hybrid_comp_mask_inverse, hybrid_comp_mask_equalize, hybrid_comp_mask_auto_contrast, hybrid_comp_save_extra_frames, hybrid_comp_alpha_schedule, hybrid_comp_mask_blend_alpha_schedule, hybrid_comp_mask_contrast_schedule, hybrid_comp_mask_auto_contrast_cutoff_high_schedule, hybrid_comp_mask_auto_contrast_cutoff_low_schedule, i1, i2, i3, i4, i5, i6, i7, i8, i9, i10, i11, i12, i13, i14, i15, i16, i17, i18, i19, i20, i21, i22, i23, i24, i25, i26, i27, i28, i29, i30, i31, i32, i33, i34, i35, i36, i37, i38, i39, i40, i41, i42) File "G:\StableDiffusion\stable-diffusion-webui\extensions\deforum\scripts\deforum.py", line 93, in run render_animation(args, anim_args, video_args, parseq_args, root.animation_prompts, root) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum/scripts\deforum_helpers\render.py", line 92, in render_animation depth_model.load_adabins(root.models_path) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum/scripts\deforum_helpers\depth.py", line 38, in load_adabins self.adabins_helper = InferenceHelper(models_path=models_path, dataset='nyu', device=self.device) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum/scripts/deforumhelpers/src\infer.py", line 85, in init model, , _ = model_io.load_checkpoint(pretrained_path, model) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum/scripts/deforum_helpers/src\model_io.py", line 37, in load_checkpoint ckpt = torch.load(fpath, map_location='cpu') File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_smartprocess\reallysafe.py", line 117, in load return load_with_extra(filename, args, kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_smartprocess\reallysafe.py", line 164, in load_with_extra return unsafe_torch_load(filename, *args, kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_dreambooth_extension\reallysafe.py", line 117, in load return load_with_extra(filename, *args, *kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_dreambooth_extension\reallysafe.py", line 164, in load_with_extra return unsafe_torch_load(filename, args, kwargs) File "G:\StableDiffusion\stable-diffusion-webui\modules\safe.py", line 106, in load return load_with_extra(filename, extra_handler=global_extra_handler, *args, *kwargs) File "G:\StableDiffusion\stable-diffusion-webui\modules\safe.py", line 151, in load_with_extra return unsafe_torch_load(filename, args, **kwargs) File "G:\StableDiffusion\stable-diffusion-webui\venv\lib\site-packages\torch\serialization.py", line 705, in load with _open_zipfile_reader(opened_file) as opened_zipfile: File "G:\StableDiffusion\stable-diffusion-webui\venv\lib\site-packages\torch\serialization.py", line 242, in init super(_open_zipfile_reader, self).init(torch._C.PyTorchFileReader(name_or_buffer)) RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

@rethink-studios and updates?

Still broken as of today's git pull on both webUI and Deforum:

To create a public link, set share=True in launch(). Deforum script for 2D, pseudo-2D and 3D animations v0.5-webui-beta Additional models path: G:\StableDiffusion\stable-diffusion-webui\models/Deforum Saving animation frames to G:\StableDiffusion\stable-diffusion-webui\outputs/img2img-images\Deforum Loading base model ()...Done. Removing last two layers (global_pool & classifier). Building Encoder-Decoder model..Done. Error completing request Arguments: ('task(70ykb2fzmdp9nod)', None, False, '', '3D', 120, 'wrap', '0:(0)', '0:(0)', '0:(0)', '0:(0)', '0:(10)', '0:(0)', '0:(0)', '0:(0)', False, '0:(0)', '0:(t%15)', '0:(0)', '0:(53)', '0: (0.08)', '0: (0.6)', '0: (1.0)', '0: (7)', False, '0: (25)', '0: (40)', '0: (200)', '0: (10000)', '0: (t%4294967293)', False, '0: ("Euler a")', '0: (5)', '0: (1.0)', '0: (0.1)', '0: (0.0)', True, 'Match Frame 0 LAB', 1.0, 1.0, 'perlin', 8.0, 8.0, 4, 0.5, True, 0.3, 200.0, 10000.0, 40.0, 'border', 'bicubic', False, '/content/video_in.mp4', 1, 0, -1, False, False, '/content/videoin.mp4', False, 4, False, '20220829210106', '[\n "a beautiful forest by Asher Brown Durand, trending on Artstation",\n "a beautiful portrait of a woman by Artgerm, trending on Artstation"\n]\n', '{\n "0": "(scenic countryside:1.0), (cherry:where(cos(6.28*t/10)>0, 1.8*cos(6.28*t/10), 0.001)), (strawberry:where(cos(6.28*t/10)<0, -1.8*cos(6.28*t/10), 0.001)), snow, detailed painting by greg rutkowski --neg (cherry:where(cos(6.28*t/10)<0, -1.8*cos(6.28*t/10), 0.001)), (strawberry:where(cos(6.28*t/10)>0, 1.8*cos(6.28*t/10), 0.001))",\n "60": "a beautiful (((banana))), trending on Artstation",\n "80": "a beautiful coconut --neg photo, realistic",\n "100": "a beautiful durian, trending on Artstation"\n}\n', 1024, 768, False, False, False, 0, 0, -1, 'Euler a', False, -1, 0, 0, 0, 21, 0.0, 1, False, 1, True, True, False, False, False, False, 'Deforum', '{timestring}{index}_{prompt}.png', 'iter', 1, False, False, True, 0, 'https://user-images.githubusercontent.com/14872007/195867706-d067cdc6-28cd-450b-a61e-55e25bc67010.png', False, False, False, True, 'https://www.filterforge.com/wiki/images/archive/b/b7/20080927223728%21Polygonal_gradient_thumb.jpg', 1.0, 1.0, 5.0, 1, True, 4, 'reroll', False, 12.0, 'PIL gif', 'ffmpeg', 'None', 'snowfall.mp3', False, False, 200.0, 'x0pred', '/content/drive/MyDrive/AI/StableDiffusion/2022-09/20220903000939%05d.png', '/content/drive/MyDrive/AI/StableDiffusion/content/drive/MyDrive/AI/StableDiffusion/2022-09/kabachuha/2022-09/20220903000939.mp4', False, '', True, False, 'None', False, 'None', 'Farneback', False, 'None', False, 'None', False, False, '0:(1)', '0:(0.5)', '0:(1)', '0:(100)', '0:(0)', '

Deforum v0.5-webui-beta

', 'Made by deforum.github.io, port for AUTOMATIC1111\'s webui maintained by kabachuha', '

The code for this extension: Fork of deforum for auto1111\'s webui.
Join the official Deforum Discord to share your creations and suggestions.
For general usage, see the User guide for Deforum v0.5.
For advanced animations, see the Math keyframing explanation.
Alternatively, use sd-parseq as a UI to define your animation schedules (see the Parseq section in the Keyframes tab).
framesync.xyz is also a good option, it makes compact math formulae for Deforum keyframes by selecting various waveforms.
The other site allows for making keyframes using interactive splines and Bezier curves (select Disco output format).
After the 2022-12-30 update, the default noise type is Perlin noise due to its great frame coherence improvements. If you want to use the old noise and replicate the previous settings, set the type to "uniform" in the Keyframes tab.

\n If you liked this extension, please give it a star on GitHub! 😊', '', '', '', '', '

Import settings from file

', '

Animation settings

', '

Motion parameters:

', '

2D and 3D settings

', '

Prespective flip — Low VRAM pseudo-3D mode:

', '

Coherence:

', '

3D Depth Warping:

', '

Video Input:

', '

Interpolation (turned off atm)

', '

Resume animation:

', '

Prompts

', '

animation_mode: None batches on list of prompts. (Batch mode disabled atm, only animation_prompts are working)

', '

Important change from vanilla Deforum!

', '

This script uses the built-in webui weighting settings.

', '

So if you want to use math functions as prompt weights,

', '

keep the values above zero in both parts

', '

Negative prompt part can be specified with --neg

', '

Run settings

', '

Sampling settings

', '

3D Fov settings:

', '', '', '', '

Batch settings

', '

Init settings

', '

3D settings

', '

Steps scheduling:

', '

Video output settings

', '

To enable seed schedule select seed behavior — \'schedule\'

', '\n Use an <a style=\'color:blue;\' target=\'_blank\' href=\'https://sd-parseq.web.app/deforum\'>sd-parseq manifest for your animation (leave blank to ignore).

\n Note that parseq overrides:\n

Run: seed, subseed, subseed strength.
Keyframes: generation settings (noise, strength, contrast, scale).
Keyframes: motion parameters for 2D and 3D (angle, zoom, translation, rotation, perspective flip).

\n <p">\n Parseq does not override:\n

Run: Sampler, W, H, Restore faces, tiling, highres fix, resize seed.
Keyframes: animation settings (animation mode, max_frames, border)
Keyframes: coherence (color coherence & diffusion cadence)
Keyframes: depth warping
Video output settings: all settings (including fps and max frames)

\n ', '

Anti-blur settings

', '

Perlin noise params, if selected.

', '

Hybrid Video Compositing in 2D/3D Modeby reallybigname

', '

Composite video with previous frame init image in 2D or 3D animation_mode (not for Video Input mode)
Uses your Init settings for video_init_path, extract_nth_frame, overwrite_extracted_frames
In Keyframes tab, you can also set color_coherence = \'Video Input\'
color_coherence_video_every_N_frames lets you only match every N frames
Color coherence may be used with hybrid composite off, to just use video color.
Hybrid motion may be used with hybrid composite off, to just use video motion.

Hybrid Video Schedules

The alpha schedule controls overall alpha for video mix, whether using a composite mask or not.
The hybrid_comp_mask_blend_alpha_schedule only affects the \'Blend\' hybrid_comp_mask_type.
Mask contrast schedule is from 0-255. Normal is 1. Affects all masks.
Autocontrast low/high cutoff schedules 0-100. Low 0 High 100 is full range.
(hybrid_comp_mask_auto_contrast must be enabled)

See main Deforum document for more details.', '

Hybrid Video Schedules

') {} Traceback (most recent call last): File "G:\StableDiffusion\stable-diffusion-webui\modules\call_queue.py", line 56, in f res = list(func(*args, kwargs)) File "G:\StableDiffusion\stable-diffusion-webui\modules\call_queue.py", line 37, in f res = func(*args, *kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\deforum\scripts\deforum.py", line 254, in run_deforum processed = DeforumScript.run(None, p, override_settings_with_file, custom_settings_file, animation_mode, max_frames, border, angle, zoom, translation_x, translation_y, translation_z, rotation_3d_x, rotation_3d_y, rotation_3d_z, flip_2d_perspective, perspective_flip_theta, perspective_flip_phi, perspective_flip_gamma, perspective_flip_fv, noise_schedule, strength_schedule, contrast_schedule, cfg_scale_schedule, enable_steps_scheduling, steps_schedule, fov_schedule, near_schedule, far_schedule, seed_schedule, enable_sampler_scheduling, sampler_schedule, kernel_schedule, sigma_schedule, amount_schedule, threshold_schedule, histogram_matching, color_coherence, color_coherence_video_every_N_frames, diffusion_cadence, noise_type, perlin_w, perlin_h, perlin_octaves, perlin_persistence, use_depth_warping, midas_weight, near_plane, far_plane, fov, padding_mode, sampling_mode, save_depth_maps, video_init_path, extract_nth_frame, extract_from_frame, extract_to_frame, overwrite_extracted_frames, use_mask_video, video_mask_path, interpolate_key_frames, interpolate_x_frames, resume_from_timestring, resume_timestring, prompts, animation_prompts, W, H, restore_faces, tiling, enable_hr, firstphase_width, firstphase_height, seed, sampler, seed_enable_extras, subseed, subseed_strength, seed_resize_from_w, seed_resize_from_h, steps, ddim_eta, n_batch, make_grid, grid_rows, save_settings, save_samples, display_samples, save_sample_per_step, show_sample_per_step, override_these_with_webui, batch_name, filename_format, seed_behavior, seed_iter_N, use_init, from_img2img_instead_of_link, strength_0_no_init, strength, init_image, use_mask, use_alpha_as_mask, invert_mask, overlay_mask, mask_file, mask_contrast_adjust, mask_brightness_adjust, mask_overlay_blur, fill, full_res_mask, full_res_mask_padding, reroll_blank_frames, skip_video_for_run_all, fps, output_format, ffmpeg_location, add_soundtrack, soundtrack_path, use_manual_settings, render_steps, max_video_frames, path_name_modifier, image_path, mp4_path, store_frames_in_ram, parseq_manifest, parseq_use_deltas, hybrid_generate_inputframes, hybrid_generate_human_masks, hybrid_use_first_frame_as_init_image, hybrid_motion, hybrid_flow_method, hybrid_composite, hybrid_comp_mask_type, hybrid_comp_mask_inverse, hybrid_comp_mask_equalize, hybrid_comp_mask_auto_contrast, hybrid_comp_save_extra_frames, hybrid_comp_alpha_schedule, hybrid_comp_mask_blend_alpha_schedule, hybrid_comp_mask_contrast_schedule, hybrid_comp_mask_auto_contrast_cutoff_high_schedule, hybrid_comp_mask_auto_contrast_cutoff_low_schedule, i1, i2, i3, i4, i5, i6, i7, i8, i9, i10, i11, i12, i13, i14, i15, i16, i17, i18, i19, i20, i21, i22, i23, i24, i25, i26, i27, i28, i29, i30, i31, i32, i33, i34, i35, i36, i37, i38, i39, i40, i41, i42) File "G:\StableDiffusion\stable-diffusion-webui\extensions\deforum\scripts\deforum.py", line 93, in run render_animation(args, anim_args, video_args, parseq_args, root.animation_prompts, root) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum/scripts\deforum_helpers\render.py", line 92, in render_animation depth_model.load_adabins(root.models_path) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum/scripts\deforum_helpers\depth.py", line 38, in load_adabins self.adabins_helper = InferenceHelper(models_path=models_path, dataset='nyu', device=self.device) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum/scripts/deforumhelpers/src\infer.py", line 85, in init model, , _ = model_io.load_checkpoint(pretrained_path, model) File "G:\StableDiffusion\stable-diffusion-webui/extensions/deforum/scripts/deforum_helpers/src\model_io.py", line 37, in load_checkpoint ckpt = torch.load(fpath, map_location='cpu') File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_smartprocess\reallysafe.py", line 117, in load return load_with_extra(filename, args, kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_smartprocess\reallysafe.py", line 164, in load_with_extra return unsafe_torch_load(filename, *args, kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_dreambooth_extension\reallysafe.py", line 117, in load return load_with_extra(filename, *args, *kwargs) File "G:\StableDiffusion\stable-diffusion-webui\extensions\sd_dreambooth_extension\reallysafe.py", line 164, in load_with_extra return unsafe_torch_load(filename, args, kwargs) File "G:\StableDiffusion\stable-diffusion-webui\modules\safe.py", line 106, in load return load_with_extra(filename, extra_handler=global_extra_handler, *args, *kwargs) File "G:\StableDiffusion\stable-diffusion-webui\modules\safe.py", line 151, in load_with_extra return unsafe_torch_load(filename, args, **kwargs) File "G:\StableDiffusion\stable-diffusion-webui\venv\lib\site-packages\torch\serialization.py", line 705, in load with _open_zipfile_reader(opened_file) as opened_zipfile: File "G:\StableDiffusion\stable-diffusion-webui\venv\lib\site-packages\torch\serialization.py", line 242, in init super(_open_zipfile_reader, self).init(torch._C.PyTorchFileReader(name_or_buffer)) RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

models/deforum

could you please check these files at your pc? they should have the same sizes, but I believe one of them is not downloaded correctly.

Update (from DM in discord):

The problem was a corrupted (small) AdaBins_nyu.pt file. @rethink-studios downloaded the file again from this link; https://drive.google.com/drive/folders/1nYyaQXOBjNdUJDsmJpcRpu6oE55aQoLA

He replaced the corrupted one, and 3D mode now work for him.

Closing.

deforum-art / sd-webui-deforum

Bug: 3D mode broken in Deforum? #211