Upload layer_diff_dataset/test_inp_sd.py with huggingface_hub

layer_diff_dataset/test_inp_sd.py

@@ -0,0 +1,53 @@
+from diffusers import AutoPipelineForInpainting
+from diffusers.utils import load_image
+import torch
+import os 
+from tqdm import tqdm
+import cv2
+from PIL import Image
+
+pipe = AutoPipelineForInpainting.from_pretrained("../alpha_work/diffusers/stable-diffusion-xl-1.0-inpainting_", torch_dtype=torch.float16, variant="fp16").to("cuda")
+# print('pipe',pipe)
+# StableDiffusionXLInpaintPipeline
+folder_path_0 = '../codes/Inpaint-Anything/results/0b6f9105fc'
+folder_path = '../data/video_dataset/YoutubeVOS/train/mask/0b6f9105fc'
+folder_path_ = 'YoutubeVOS/inp_preprocess_sd_0.9_base/0b6f9105fc'
+os.makedirs(folder_path_,exist_ok=True)
+file_list = os.listdir(folder_path)
+file_list = [i for i in file_list if i.endswith('.png')]
+file_list.sort()
+# img_url = "https://raw.githubusercontent.com/CompVis/latent-diffusion/main/data/inpainting_examples/overture-creations-5sI6fQgYIuo.png"
+# mask_url = "https://raw.githubusercontent.com/CompVis/latent-diffusion/main/data/inpainting_examples/overture-creations-5sI6fQgYIuo_mask.png"
+
+
+prompt = "hazy background with nothing on"
+generator = torch.Generator(device="cuda").manual_seed(0)
+# base_image = Image.open(base_image_path).resize((1024, 1024))
+
+pbar = tqdm(enumerate(file_list),total=len(file_list))
+for i, image_name in pbar:
+    # if os.path.exists(os.path.join(folder_path_,image_name)):
+    #     continue
+    image_path = os.path.join(folder_path_0,image_name)
+    mask_path = os.path.join(folder_path,image_name)
+    image = Image.open(image_path).resize((1024, 1024))
+    mask_image = Image.open(mask_path).resize((1024, 1024))
+    # image = cv2.resize(cv2.imread(image_path),(1024,1024))
+    # mask_image = cv2.resize(cv2.imread(mask_path,cv2.IMREAD_GRAYSCALE),(1024,1024))
+    # image = load_image(img_url).resize((1024, 1024))
+    # mask_image = load_image(mask_url).resize((1024, 1024))
+    if i==0:
+        base_image = image
+    image_out = pipe(
+    prompt=prompt,
+    image=image,
+    base_image=base_image,
+    mask_image=mask_image,
+    guidance_scale=8.0,
+    num_inference_steps=20,  # steps between 15 and 30 work well for us
+    strength=0.9,  # make sure to use `strength` below 1.0
+    generator=generator,
+    ).images[0]
+    image_out.save(os.path.join(folder_path_,image_name))
+    if i==0:
+        base_image = image_out