search

Tok / sd-forge-deforum

Fluxabled fork of the Deforum extension for Stable Dissusion WebUI Forge with keyframe redistribution logic.
Other
36 stars 1 forks source link

[Bug]: openCV error -215 while using Parseq #12

Open furibrz opened 1 month ago

furibrz commented 1 month ago

What happened and why was it ungood?

Not sure whether to post this here or to Parseq repo. While trying to generate a video using default parseq document:

Info: Sampler: 'DPM++ 2M SDE' Scheduler: 'Karras'

Warning: Frames have been removed. Original count: 19, New count: 8

Warning: Frames have been removed. Original count: 19, New count: 8

Warning: Frames have been removed. Original count: 19, New count: 8

Warning: Frames have been removed. Original count: 19, New count: 8

Info: Calculated pseudo cadence: 30.25 Info: Precalculating 4 seeds with behaviour 'iter'. Info: Created 4 key frames with 120 tweens. Key frame index distribution: 'KEYFRAMES_ONLY'.

Animation frame: 1/121 Seed: -1 Prompt: A lone (black cat:0.89117) (white duck:0.10883) at midday, centered, realistic, photorealism, crisp, natural colors, fine textures, highly detailed, volumetric lighting, studio photography Neg Prompt: (black cat:0.10883) (white duck:0.89117) watermark, logo, text, signature, copyright, writing, letters, low quality, artefacts, cropped, bad art, poorly drawn, lowres, simple, pixelated, grain, noise, blurry, cartoon, computer game, video game, painting, drawing, sketch, disfigured, deformed, ugly Not using an init image (doing pure txt2img) ╭─────┬───┬─────────┬───────┬───────┬─────────┬────┬────┬────┬────┬────┬────┬──────────╮ │Steps│CFG│Dist. CFG│Denoise│Subseed│Subs. str│Tr X│Tr Y│Tr Z│Ro X│Ro Y│Ro Z│Asp. Ratio│ ├─────┼───┼─────────┼───────┼───────┼─────────┼────┼────┼────┼────┼────┼────┼──────────┤ │ 8 │2.5│ 2.5 │ 0 │ -1 │ 0 │ 0 │ 0 │1.1 │ 0 │ 0 │ 0 │ 1 │ ╰─────┴───┴─────────┴───────┴───────┴─────────┴────┴────┴────┴────┴────┴────┴──────────╯ [Unload] Trying to free 2457.60 MB for cuda:0 with 1 models keep loaded ... Current free memory is 14643.73 MB ... Done. 100%|######################################################################################################################################| 8/8 [00:03<00:00, 2.30it/s] Tweens progress: 0%| | 0/40 [00:00<?, ?it/s] START OF TRACEBACK Traceback (most recent call last): File "C:\AI\forge\webui\extensions\sd-forge-deforum\scripts\deforum_helpers\run_deforum.py", line 124, in run_deforum render_animation(args, anim_args, video_args, parseq_args, loop_args, controlnet_args, freeu_args, kohya_hrfix_args, root) File "C:\AI\forge\webui\extensions\sd-forge-deforum\scripts\deforum_helpers\render.py", line 64, in render_animation experimental_core.render_animation(args, anim_args, video_args, parseq_args, loop_args, controlnet_args, freeu_args, kohya_hrfix_args, root) File "C:\AI\forge\webui\extensions\sd-forge-deforum\scripts\deforum_helpers\rendering\experimental_core.py", line 26, in render_animation run_render_animation(data, diffusion_frames) File "C:\AI\forge\webui\extensions\sd-forge-deforum\scripts\deforum_helpers\rendering\experimental_core.py", line 34, in run_render_animation _pre_process_diffusion_frame_and_emit_tweens(data, diffusion_frame) File "C:\AI\forge\webui\extensions\sd-forge-deforum\scripts\deforum_helpers\rendering\experimental_core.py", line 46, in _pre_process_diffusion_frame_and_emit_tweens emit_tweens(data, diffusion_frame) File "C:\AI\forge\webui\extensions\sd-forge-deforum\scripts\deforum_helpers\rendering\experimental_core.py", line 67, in emit_tweens [tween.emit_frame(frame, grayscale_tube, overlay_mask_tube) for tween in tweens] File "C:\AI\forge\webui\extensions\sd-forge-deforum\scripts\deforum_helpers\rendering\experimental_core.py", line 67, in [tween.emit_frame(frame, grayscale_tube, overlay_mask_tube) for tween in tweens] File "C:\AI\forge\webui\extensions\sd-forge-deforum\scripts\deforum_helpers\rendering\data\frame\tween_frame.py", line 36, in emit_frame new_image = image_utils.save_and_return_frame(data, self, new_image) File "C:\AI\forge\webui\extensions\sd-forge-deforum\scripts\deforum_helpers\rendering\util\image_utils.py", line 40, in save_and_return_frame save_cadence_frame_and_depth_map_if_active(data, frame, image) File "C:\AI\forge\webui\extensions\sd-forge-deforum\scripts\deforum_helpers\rendering\util\image_utils.py", line 33, in save_cadence_frame_and_depth_map_if_active save_cadence_frame(data, frame.i(), image) File "C:\AI\forge\webui\extensions\sd-forge-deforum\scripts\deforum_helpers\rendering\util\image_utils.py", line 29, in save_cadence_frame cv2.imwrite(save_path, image) cv2.error: OpenCV(4.10.0) D:\a\opencv-python\opencv-python\opencv\modules\imgcodecs\src\loadsave.cpp:798: error: (-215:Assertion failed) !_img.empty() in function 'cv::imwrite'

END OF TRACEBACK

User friendly error message: Error: OpenCV(4.10.0) D:\a\opencv-python\opencv-python\opencv\modules\imgcodecs\src\loadsave.cpp:798: error: (-215:Assertion failed) !_img.empty() in function 'cv::imwrite' . Please, check your schedules/ init values. Tweens progress: 0%|

asdrabael commented 1 month ago

Do you get the same error if you use a different scheduler? Also what model? That error indicates that something is causing an error where it's not saving pictures. 2.5 cfg and 2.5 distilled cfg is also something I've never seem someone do. If you're using Flux some schedulers give funky errors.

furibrz commented 1 month ago

Hey. I'm using the DreamShaper XL Turbo model, hence the low cfg and steps. i'm using the DPM++ sampler with the Karras scheduler. any other scheduler + sampler combo gives me the same error.

asdrabael commented 1 month ago

I'm not sure if it works with turbo models. That's a distinction outside of my expertise. Some stuff like controlnet gives similar issues, so that's my best guess until Tok weighs in