Upload 4 files

app.py

@@ -0,0 +1,93 @@
+import spaces
+import gradio as gr
+import torch
+from diffusers import (
+    AutoencoderKL,
+    EulerAncestralDiscreteScheduler,
+)
+from diffusers.utils import load_image
+from replace_bg.model.pipeline_controlnet_sd_xl import StableDiffusionXLControlNetPipeline
+from replace_bg.model.controlnet import ControlNetModel
+from replace_bg.utilities import resize_image, remove_bg_from_image, paste_fg_over_image, get_control_image_tensor
+
+controlnet = ControlNetModel.from_pretrained("briaai/BRIA-2.3-ControlNet-BG-Gen", torch_dtype=torch.float16) 
+vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16)
+pipe = StableDiffusionXLControlNetPipeline.from_pretrained("briaai/BRIA-2.3", controlnet=controlnet, torch_dtype=torch.float16, vae=vae).to('cuda:0')
+pipe.scheduler = EulerAncestralDiscreteScheduler(
+    beta_start=0.00085,
+    beta_end=0.012,
+    beta_schedule="scaled_linear",
+    num_train_timesteps=1000,
+    steps_offset=1
+)
+
+
+@spaces.GPU
+def generate_(prompt, negative_prompt, control_tensor, num_steps, controlnet_conditioning_scale, seed):
+    generator = torch.Generator("cuda").manual_seed(seed)    
+    gen_img = pipe(
+        negative_prompt=negative_prompt, 
+        prompt=prompt,     
+        controlnet_conditioning_scale=float(controlnet_conditioning_scale), 
+        num_inference_steps=num_steps,
+        image = control_tensor,
+        generator=generator
+    ).images[0]
+    
+    return gen_img
+
+@spaces.GPU
+def process(input_image, prompt, negative_prompt, num_steps, controlnet_conditioning_scale, seed):
+    
+    image = resize_image(input_image)
+    mask = remove_bg_from_image(image)
+    control_tensor = get_control_image_tensor(pipe.vae, image, mask)    
+  
+    gen_image = generate_(prompt, negative_prompt, control_tensor, num_steps, controlnet_conditioning_scale, seed)
+    result_image = paste_fg_over_image(gen_image, image, mask)
+
+    return result_image
+
+
+
+block = gr.Blocks().queue()
+
+with block:
+    gr.Markdown("## HBS_V2")
+    gr.HTML('''
+      <p style="margin-bottom: 10px; font-size: 94%">
+      Human Body Segmentation model v2 developed by <a href='https://github.com/WildanJR09' target='_blank'><b>WildanJR</b></a>, Designed to effectively separate foreground from background in a range of categories and image types. And then generate image background from user input.<br> 
+This model has been trained on a carefully selected dataset, which includes: general stock images, e-commerce, gaming, and advertising content, making it suitable for commercial use cases powering enterprise content creation at scale. The accuracy, efficiency, and versatility currently rival leading source-available models. It is ideal where content safety, legally licensed datasets, and bias mitigation are paramount. For test upload your image and type query then wait.
+      </p>
+    ''')
+    with gr.Row():
+        with gr.Column():
+            input_image = gr.Image(sources='upload', type="pil", label="Upload", elem_id="image_upload", height=600) 
+            prompt = gr.Textbox(label="Prompt for Background")
+            run_button = gr.Button(value="Generate Background")
+
+        with gr.Column():
+            result_gallery = gr.Image(label='Output', type="pil", show_label=True, elem_id="output-img")
+
+    # Define default values for hidden parameters
+    negative_prompt = "Logo,Watermark,Text,Ugly,Morbid,Extra fingers,Poorly drawn hands,Mutation,Blurry,Extra limbs,Gross proportions,Missing arms,Mutated hands,Long neck,Duplicate,Mutilated,Mutilated hands,Poorly drawn face,Deformed,Bad anatomy,Cloned face,Malformed limbs,Missing legs,Too many fingers"
+    num_steps = 30
+    controlnet_conditioning_scale = 1.0
+    seed = 309005275  # None for random seed, or specify a fixed integer
+            
+ 
+            # result_gallery = gr.Gallery(label='Output', show_label=False, elem_id="gallery", columns=[1], height=600)
+    ips = [input_image, prompt, negative_prompt, num_steps, controlnet_conditioning_scale, seed]
+    run_button.click(fn=process, inputs=ips, outputs=[result_gallery])
+
+    gr.Examples(
+                examples=[
+                    ["./jisoo.png"],
+                ],
+                fn=process,
+                inputs=[input_image],
+                cache_examples=False,
+    )
+
+
+block.launch(debug = True)

requirements.txt

@@ -0,0 +1,9 @@
+torch
+torchvision
+pillow
+numpy
+typing
+scikit-image
+Diffusers==0.26.2
+transformers>=4.39.1
+accelerate