不支持识别中文路径吗？

[qq14909]

iopaint run --model=lama --device=cpu --image=E:\水印图片 --mask=E:\水印图片\Mask.png --output=E:\去除水印后
2024-03-08 18:21:03.632 | INFO | iopaint.runtime:setup_model_dir:82 - Model directory: C:\Users\Administrator.cache
2024-03-08 18:21:06.277 | INFO | iopaint.batch_processing:batch_inpaint:65 - Using default config: image=None mask=None ldm_steps=20 ldm_sampler=<LDMSampler.plms: 'plms'> zits_wireframe=True hd_strategy=<HDStrategy.CROP: 'Crop'> hd_strategy_crop_trigger_size=800 hd_strategy_crop_margin=128 hd_strategy_resize_limit=1280 prompt='' negative_prompt='' use_croper=False croper_x=0 croper_y=0 croper_height=512 croper_width=512 use_extender=False extender_x=0 extender_y=0 extender_height=640 extender_width=640 sd_scale=1.0 sd_mask_blur=11 sd_strength=1.0 sd_steps=50 sd_guidance_scale=7.5 sd_sampler=<SDSampler.uni_pc: 'UniPC'> sd_seed=42 sd_match_histograms=False sd_outpainting_softness=20.0 sd_outpainting_space=20.0 sd_freeu=False sd_freeu_config=FREEUConfig(s1=0.9, s2=0.2, b1=1.2, b2=1.4) sd_lcm_lora=False sd_keep_unmasked_area=True cv2_flag=<CV2Flag.INPAINT_NS: 'INPAINT_NS'> cv2_radius=4 paint_by_example_example_image=None p2p_image_guidance_scale=1.5 enable_controlnet=False controlnet_conditioning_scale=0.4 controlnet_method='lllyasviel/control_v11p_sd15_canny' powerpaint_task=<PowerPaintTask.text_guided: 'text-guided'> fitting_degree=1.0
2024-03-08 18:21:06.278 | INFO | iopaint.model_manager:init_model:38 - Loading model: lama
2024-03-08 18:21:06.279 | INFO | iopaint.helper:load_jit_model:107 - Loading model from: C:\Users\Administrator.cache\torch\hub\checkpoints\big-lama.pt
⠋ Batch processing... ---------------------------------------- 0% 0/6 0:00:00[ WARN:[email protected]] global loadsave.cpp:248 cv::findDecoder imread_('E:\姘村嵃鍥剧墖\082826baqc53xr35xza45u.jpg'): can't open/read file: check file path/integrity
⠋ Batch processing... ---------------------------------------- 0% 0/6 0:00:00
┌─────────────────────────────── Traceback (most recent call last) ────────────────────────────────┐
│ C:\Program Files\Python312\Lib\site-packages\iopaint\cli.py:94 in run │
│ │
│ 91 │ │
│ 92 │ from iopaint.batch_processing import batch_inpaint │
│ 93 │ │
│ > 94 │ batch_inpaint(model, device, image, mask, output, config, concat) │
│ 95 │
│ 96 │
│ 97 @typer_app.command(help="Start IOPaint server") │
│ │
│ C:\Program Files\Python312\Lib\site-packages\iopaint\batch_processing.py:96 in batch_inpaint │
│ │
│ 93 │ │ │ infos = Image.open(image_p).info │
│ 94 │ │ │ │
│ 95 │ │ │ img = cv2.imread(str(image_p)) │
│ > 96 │ │ │ img = cv2.cvtColor(img, cv2.COLOR_BGRA2RGB) │
│ 97 │ │ │ mask_img = cv2.imread(str(mask_p), cv2.IMREAD_GRAYSCALE) │
│ 98 │ │ │ if mask_img.shape[:2] != img.shape[:2]: │
│ 99 │ │ │ │ progress.log( │
└──────────────────────────────────────────────────────────────────────────────────────────────────┘
error: OpenCV(4.9.0) D:\a\opencv-python\opencv-python\opencv\modules\imgproc\src\color.cpp:196: error: (-215:Assertion failed) !_src.empty() in function 'cv::cvtColor'

[yumianhuli1]

是的

[waylake]

The path problem of the file being read by opencv.
If there are Chinese, Korean, or Japanese in the path of the image, an error occurs.
In my case, I solved it through a relative path that does not include the above language, but it is fundamentally a path problem.