high quality (#17)

- High quality (0f59bfafcd98b4b4a93790a81be7de43d068135f)
- Update app.py (09d2a3966a11c5d694bde9833a8a514a0e291a14)


Co-authored-by: Apolinário from multimodal AI art <[email protected]>

app.py

@@ -8,6 +8,8 @@ from diffusers import (
     StableDiffusionControlNetPipeline,
     ControlNetModel,
     StableDiffusionLatentUpscalePipeline,
+    StableDiffusionImg2ImgPipeline,
+    StableDiffusionControlNetImg2ImgPipeline,
     DPMSolverMultistepScheduler,  # <-- Added import
     EulerDiscreteScheduler  # <-- Added import
 )
@@ -18,17 +20,22 @@ from illusion_style import css
 BASE_MODEL = "SG161222/Realistic_Vision_V5.1_noVAE"
 
 # Initialize both pipelines
-vae = AutoencoderKL.from_pretrained("stabilityai/sd-vae-ft-mse")
+vae = AutoencoderKL.from_pretrained("stabilityai/sd-vae-ft-mse", torch_dtype=torch.float16)
 #init_pipe = DiffusionPipeline.from_pretrained("SG161222/Realistic_Vision_V5.1_noVAE", torch_dtype=torch.float16)
-controlnet = ControlNetModel.from_pretrained("monster-labs/control_v1p_sd15_qrcode_monster")#, torch_dtype=torch.float16)
+controlnet = ControlNetModel.from_pretrained("monster-labs/control_v1p_sd15_qrcode_monster", torch_dtype=torch.float16)#, torch_dtype=torch.float16)
 main_pipe = StableDiffusionControlNetPipeline.from_pretrained(
     BASE_MODEL,
     controlnet=controlnet,
     vae=vae,
     safety_checker=None,
-    #torch_dtype=torch.float16,
+    torch_dtype=torch.float16,
 ).to("cuda")
+#main_pipe.unet = torch.compile(main_pipe.unet, mode="reduce-overhead", fullgraph=True)
+#main_pipe.unet.to(memory_format=torch.channels_last)
+#main_pipe.unet = torch.compile(main_pipe.unet, mode="reduce-overhead", fullgraph=True)
 #model_id = "stabilityai/sd-x2-latent-upscaler"
+image_pipe = StableDiffusionControlNetImg2ImgPipeline.from_pretrained(BASE_MODEL, unet=main_pipe.unet, vae=vae, controlnet=controlnet, safety_checker=None, torch_dtype=torch.float16).to("cuda")
+#image_pipe.unet = torch.compile(image_pipe.unet, mode="reduce-overhead", fullgraph=True)
 #upscaler = StableDiffusionLatentUpscalePipeline.from_pretrained(model_id, torch_dtype=torch.float16)
 #upscaler.to("cuda")
 
@@ -55,6 +62,31 @@ def center_crop_resize(img, output_size=(512, 512)):
 
     return img
 
+def common_upscale(samples, width, height, upscale_method, crop=False):
+        if crop == "center":
+            old_width = samples.shape[3]
+            old_height = samples.shape[2]
+            old_aspect = old_width / old_height
+            new_aspect = width / height
+            x = 0
+            y = 0
+            if old_aspect > new_aspect:
+                x = round((old_width - old_width * (new_aspect / old_aspect)) / 2)
+            elif old_aspect < new_aspect:
+                y = round((old_height - old_height * (old_aspect / new_aspect)) / 2)
+            s = samples[:,:,y:old_height-y,x:old_width-x]
+        else:
+            s = samples
+
+        return torch.nn.functional.interpolate(s, size=(height, width), mode=upscale_method)
+
+def upscale(samples, upscale_method, scale_by):
+        #s = samples.copy()
+        width = round(samples["images"].shape[3] * scale_by)
+        height = round(samples["images"].shape[2] * scale_by)
+        s = common_upscale(samples["images"], width, height, upscale_method, "disabled")
+        return (s)
+
 # Inference function
 def inference(
     control_image: Image.Image,
@@ -62,6 +94,9 @@ def inference(
     negative_prompt: str,
     guidance_scale: float = 8.0,
     controlnet_conditioning_scale: float = 1,
+    control_guidance_start: float = 1,    
+    control_guidance_end: float = 1,
+    upscaler_strength: float = 0.5,
     seed: int = -1,
     sampler = "DPM++ Karras SDE",
     progress = gr.Progress(track_tqdm=True)
@@ -73,65 +108,82 @@ def inference(
     #init_image = init_pipe(prompt).images[0]
 
     # Rest of your existing code
-    control_image = center_crop_resize(control_image)
+    control_image_small = center_crop_resize(control_image)
     main_pipe.scheduler = SAMPLER_MAP[sampler](main_pipe.scheduler.config)
     generator = torch.manual_seed(seed) if seed != -1 else torch.Generator()
 
     out = main_pipe(
         prompt=prompt,
         negative_prompt=negative_prompt,
-        image=control_image,
-        #control_image=control_image,
+        image=control_image_small,
         guidance_scale=float(guidance_scale),
         controlnet_conditioning_scale=float(controlnet_conditioning_scale),
         generator=generator,
-        #strength=strength,
-        num_inference_steps=30,
-        #output_type="latent"
-    ).images[0]
-    
-    return out, gr.update(visible=True)
+        control_guidance_start=float(control_guidance_start),
+        control_guidance_end=float(control_guidance_end),
+        num_inference_steps=15,
+        output_type="latent"
+    )
+    control_image_large = center_crop_resize(control_image, (1024, 1024))
+    upscaled_latents = upscale(out, "nearest-exact", 2)
+    out_image = image_pipe(
+        prompt=prompt,
+        negative_prompt=negative_prompt,
+        control_image=control_image_large,        
+        image=upscaled_latents,
+        guidance_scale=float(guidance_scale),
+        generator=generator,
+        num_inference_steps=20,
+        strength=upscaler_strength,
+        control_guidance_start=float(control_guidance_start),
+        control_guidance_end=float(control_guidance_end),
+        controlnet_conditioning_scale=float(controlnet_conditioning_scale)
+    )
+    return out_image["images"][0], gr.update(visible=True)
+        
+    #return out
 
 with gr.Blocks(css=css) as app:
     gr.Markdown(
         '''
-        <center><h1>Illusion Diffusion 🌀</h1></span>  
-        <span font-size:16px;">Generate stunning illusion artwork with Stable Diffusion</span>  
+        <center><h1>Illusion Diffusion HQ 🌀</h1></span>  
+        <span font-size:16px;">Generate stunning high quality illusion artwork with Stable Diffusion</span>  
         </center>
  
         A space by AP [Follow me on Twitter](https://twitter.com/angrypenguinPNG)
 
         This project works by using [Monster Labs QR Control Net](https://huggingface.co/monster-labs/control_v1p_sd15_qrcode_monster).
-        Given a prompt and your pattern, we use a QR code conditioned controlnet to create a stunning illusion! Credit to: MrUgleh (https://twitter.com/MrUgleh) for discovering the workflow :)
-
+        Given a prompt and your pattern, we use a QR code conditioned controlnet to create a stunning illusion! Credit to: [MrUgleh](https://twitter.com/MrUgleh) for discovering the workflow :)
         '''
     )
     
     with gr.Row():
         with gr.Column():
             control_image = gr.Image(label="Input Illusion", type="pil", elem_id="control_image")
-            controlnet_conditioning_scale = gr.Slider(minimum=0.0, maximum=5.0, step=0.01, value=0.8, label="Illusion strength", info="ControlNet conditioning scale", elem_id="illusion_strength")
-            gr.Examples(examples=["checkers.png", "pattern.png", "spiral.jpeg"], inputs=control_image)
+            controlnet_conditioning_scale = gr.Slider(minimum=0.0, maximum=5.0, step=0.01, value=0.8, label="Illusion strength", elem_id="illusion_strength", info="ControlNet conditioning scale")
+            gr.Examples(examples=["checkers.png", "checkers_mid.jpg", "pattern.png", "ultra_checkers.png", "spiral.jpeg", "funky.jpeg" ], inputs=control_image)
             prompt = gr.Textbox(label="Prompt", elem_id="prompt")
             negative_prompt = gr.Textbox(label="Negative Prompt", value="low quality", elem_id="negative_prompt")
             with gr.Accordion(label="Advanced Options", open=False):
-                #strength = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, value=0.9, label="Strength")
                 guidance_scale = gr.Slider(minimum=0.0, maximum=50.0, step=0.25, value=7.5, label="Guidance Scale")
                 sampler = gr.Dropdown(choices=list(SAMPLER_MAP.keys()), value="Euler")
-                seed = gr.Slider(minimum=-1, maximum=9999999999, step=1, value=2313123, label="Seed", randomize=True)
+                control_start = gr.Slider(minimum=0.0, maximum=1.0, step=0.1, value=0, label="Start of ControlNet")
+                control_end = gr.Slider(minimum=0.0, maximum=1.0, step=0.1, value=1, label="End of ControlNet")
+                strength = gr.Slider(minimum=0.0, maximum=1.0, step=0.1, value=1, label="Strength of the upscaler")
+                seed = gr.Slider(minimum=-1, maximum=9999999999, step=1, value=-1, label="Seed", info="-1 means random seed", randomize=True)
             run_btn = gr.Button("Run")
         with gr.Column():
-            result_image = gr.Image(label="Illusion Diffusion Output", elem_id="output")
+            result_image = gr.Image(label="Illusion Diffusion Output", interactive=False, elem_id="output")
             with gr.Group(elem_id="share-btn-container", visible=False) as share_group:
                 community_icon = gr.HTML(community_icon_html)
                 loading_icon = gr.HTML(loading_icon_html)
                 share_button = gr.Button("Share to community", elem_id="share-btn")
 
     history = show_gallery_history()
-
+    
     run_btn.click(
         inference,
-        inputs=[control_image, prompt, negative_prompt, guidance_scale, controlnet_conditioning_scale, seed, sampler],
+        inputs=[control_image, prompt, negative_prompt, guidance_scale, controlnet_conditioning_scale, control_start, control_end, strength, seed, sampler],
         outputs=[result_image, share_group]
     ).then(
         fn=fetch_gallery_history, inputs=[prompt, result_image], outputs=history, queue=False