Update app.py

app.py

@@ -1,7 +1,6 @@
 import gradio as gr
 import spaces
 from gradio_litmodel3d import LitModel3D
-
 import os
 import shutil
 os.environ['SPCONV_ALGO'] = 'native'
@@ -14,26 +13,11 @@ from PIL import Image, ImageOps
 from trellis.pipelines import TrellisImageTo3DPipeline
 from trellis.representations import Gaussian, MeshExtractResult
 from trellis.utils import render_utils, postprocessing_utils
-
-import os
-import random
 import torch
 import torchvision.transforms.functional as TF
-
 from diffusers import ControlNetModel, StableDiffusionXLControlNetPipeline, AutoencoderKL
 from diffusers import DDIMScheduler, EulerAncestralDiscreteScheduler
-from controlnet_aux import PidiNetDetector, HEDdetector
-from diffusers.utils import load_image
-from huggingface_hub import HfApi
 from pathlib import Path
-from PIL import Image, ImageOps
-import torch
-import numpy as np
-import cv2
-import os
-import random
-from gradio_imageslider import ImageSlider
-
 style_list = [
     {
         "name": "(No style)",
@@ -59,7 +43,6 @@ style_list = [
 styles = {k["name"]: (k["prompt"], k["negative_prompt"]) for k in style_list}
 STYLE_NAMES = list(styles.keys())
 DEFAULT_STYLE_NAME = "(No style)"
-
 MAX_SEED = np.iinfo(np.int32).max
 TMP_DIR = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'tmp')
 os.makedirs(TMP_DIR, exist_ok=True)
@@ -74,7 +57,7 @@ def apply_style(style_name: str, positive: str, negative: str = "") -> tuple[str
 def start_session(req: gr.Request):
     user_dir = os.path.join(TMP_DIR, str(req.session_hash))
     os.makedirs(user_dir, exist_ok=True)
-    
+
 def end_session(req: gr.Request):
     user_dir = os.path.join(TMP_DIR, str(req.session_hash))
     shutil.rmtree(user_dir)
@@ -93,11 +76,8 @@ def preprocess_image(image: Image.Image,
     new_width, new_height = int(width * ratio), int(height * ratio)
     image = image['composite'].resize((new_width, new_height))
     image = ImageOps.invert(image)
-
     print("image:",type(image))
-
     prompt, negative_prompt = apply_style(style_name, prompt, negative_prompt)
-    
     print("params:", prompt, negative_prompt, style_name, num_steps, guidance_scale, controlnet_conditioning_scale)
     output = pipe_control(
         prompt=prompt,
@@ -108,15 +88,9 @@ def preprocess_image(image: Image.Image,
         guidance_scale=guidance_scale,
         width=new_width,
         height=new_height).images[0]
-    
     processed_image = pipeline.preprocess_image(output)
     return (image, processed_image)
 
-def preprocess_images(images: List[Tuple[Image.Image, str]]) -> List[Image.Image]:
-    images = [image[0] for image in images]
-    processed_images = [pipeline.preprocess_image(image) for image in images]
-    return processed_images
-
 def pack_state(gs: Gaussian, mesh: MeshExtractResult) -> dict:
     return {
         'gaussian': {
@@ -132,7 +106,7 @@ def pack_state(gs: Gaussian, mesh: MeshExtractResult) -> dict:
             'faces': mesh.faces.cpu().numpy(),
         },
     }
-    
+
 def unpack_state(state: dict) -> Tuple[Gaussian, edict, str]:
     gs = Gaussian(
         aabb=state['gaussian']['aabb'],
@@ -147,12 +121,10 @@ def unpack_state(state: dict) -> Tuple[Gaussian, edict, str]:
     gs._scaling = torch.tensor(state['gaussian']['_scaling'], device='cuda')
     gs._rotation = torch.tensor(state['gaussian']['_rotation'], device='cuda')
     gs._opacity = torch.tensor(state['gaussian']['_opacity'], device='cuda')
-    
     mesh = edict(
         vertices=torch.tensor(state['mesh']['vertices'], device='cuda'),
         faces=torch.tensor(state['mesh']['faces'], device='cuda'),
     )
-    
     return gs, mesh
 
 def get_seed(randomize_seed: bool, seed: int) -> int:
@@ -161,48 +133,28 @@ def get_seed(randomize_seed: bool, seed: int) -> int:
 @spaces.GPU
 def image_to_3d(
     image: Image.Image,
-    multiimages: List[Tuple[Image.Image, str]],
-    is_multiimage: bool,
     seed: int,
     ss_guidance_strength: float,
     ss_sampling_steps: int,
     slat_guidance_strength: float,
     slat_sampling_steps: int,
-    multiimage_algo: Literal["multidiffusion", "stochastic"],
     req: gr.Request,
 ) -> Tuple[dict, str]:
     user_dir = os.path.join(TMP_DIR, str(req.session_hash))
-    if not is_multiimage:
-        outputs = pipeline.run(
-            image[1],
-            seed=seed,
-            formats=["gaussian", "mesh"],
-            preprocess_image=False,
-            sparse_structure_sampler_params={
-                "steps": ss_sampling_steps,
-                "cfg_strength": ss_guidance_strength,
-            },
-            slat_sampler_params={
-                "steps": slat_sampling_steps,
-                "cfg_strength": slat_guidance_strength,
-            },
-        )
-    else:
-        outputs = pipeline.run_multi_image(
-            [image[0] for image in multiimages],
-            seed=seed,
-            formats=["gaussian", "mesh"],
-            preprocess_image=False,
-            sparse_structure_sampler_params={
-                "steps": ss_sampling_steps,
-                "cfg_strength": ss_guidance_strength,
-            },
-            slat_sampler_params={
-                "steps": slat_sampling_steps,
-                "cfg_strength": slat_guidance_strength,
-            },
-            mode=multiimage_algo,
-        )
+    outputs = pipeline.run(
+        image[1],
+        seed=seed,
+        formats=["gaussian", "mesh"],
+        preprocess_image=False,
+        sparse_structure_sampler_params={
+            "steps": ss_sampling_steps,
+            "cfg_strength": ss_guidance_strength,
+        },
+        slat_sampler_params={
+            "steps": slat_sampling_steps,
+            "cfg_strength": slat_guidance_strength,
+        },
+    )
     video = render_utils.render_video(outputs['gaussian'][0], num_frames=120)['color']
     video_geo = render_utils.render_video(outputs['mesh'][0], num_frames=120)['normal']
     video = [np.concatenate([video[i], video_geo[i]], axis=1) for i in range(len(video))]
@@ -227,10 +179,6 @@ def extract_glb(
     torch.cuda.empty_cache()
     return glb_path, glb_path
 
-def reset_do_preprocess():
-    return True
-    
-@spaces.GPU
 def extract_gaussian(state: dict, req: gr.Request) -> Tuple[str, str]:
     user_dir = os.path.join(TMP_DIR, str(req.session_hash))
     gs, _ = unpack_state(state)
@@ -239,30 +187,6 @@ def extract_gaussian(state: dict, req: gr.Request) -> Tuple[str, str]:
     torch.cuda.empty_cache()
     return gaussian_path, gaussian_path
 
-def prepare_multi_example() -> List[Image.Image]:
-    multi_case = list(set([i.split('_')[0] for i in os.listdir("assets/example_multi_image")]))
-    images = []
-    for case in multi_case:
-        _images = []
-        for i in range(1, 4):
-            img = Image.open(f'assets/example_multi_image/{case}_{i}.png')
-            W, H = img.size
-            img = img.resize((int(W / H * 512), 512))
-            _images.append(np.array(img))
-        images.append(Image.fromarray(np.concatenate(_images, axis=1)))
-    return images
-
-def split_image(image: Image.Image) -> List[Image.Image]:
-    image = np.array(image)
-    alpha = image[..., 3]
-    alpha = np.any(alpha>0, axis=0)
-    start_pos = np.where(~alpha[:-1] & alpha[1:])[0].tolist()
-    end_pos = np.where(alpha[:-1] & ~alpha[1:])[0].tolist()
-    images = []
-    for s, e in zip(start_pos, end_pos):
-        images.append(Image.fromarray(image[:, s:e+1]))
-    return [preprocess_image(image) for image in images]
-
 with gr.Blocks(delete_cache=(600, 600)) as demo:
     gr.Markdown("""
     # UTPL - Conversión de Boceto a objetos 3D usando IA
@@ -271,7 +195,6 @@ with gr.Blocks(delete_cache=(600, 600)) as demo:
     **Base técnica:** Adaptación de TRELLIS (herramienta de código abierto para generación 3D)
     **Propósito educativo:** Demostraciones académicas e Investigación en modelado 3D automático
     """)
-    
     with gr.Row():
         with gr.Column():
             with gr.Column():    
@@ -282,11 +205,9 @@ with gr.Blocks(delete_cache=(600, 600)) as demo:
                 with gr.Row():
                     prompt = gr.Textbox(label="Prompt")
                     style = gr.Dropdown(label="Style", choices=STYLE_NAMES, value=DEFAULT_STYLE_NAME)
-
                 with gr.Accordion(label="Generation Settings", open=False):
                     with gr.Tab(label="sketch-to-image generation"):
                         negative_prompt = gr.Textbox(label="Negative prompt")
-                        
                         num_steps = gr.Slider(
                         label="Number of steps",
                         minimum=1,
@@ -319,71 +240,31 @@ with gr.Blocks(delete_cache=(600, 600)) as demo:
                         with gr.Row():
                             slat_guidance_strength = gr.Slider(0.0, 10.0, label="Guidance Strength", value=3.0, step=0.1)
                             slat_sampling_steps = gr.Slider(1, 50, label="Sampling Steps", value=12, step=1)
-                        multiimage_algo = gr.Radio(["stochastic", "multidiffusion"], label="Multi-image Algorithm", value="stochastic")
-                   
-            with gr.Tab(label="Multiple Images", id=1, visible=False) as multiimage_input_tab:
-                    multiimage_prompt = gr.Gallery(label="Image Prompt", format="png", type="pil", height=300, columns=3)
-                    gr.Markdown("""
-                        Input different views of the object in separate images. 
-                        
-                        *NOTE: this is an experimental algorithm without training a specialized model. It may not produce the best results for all images, especially those having different poses or inconsistent details.*
-                    """)
-            
             with gr.Accordion(label="GLB Extraction Settings", open=False):
                 mesh_simplify = gr.Slider(0.9, 0.98, label="Simplify", value=0.95, step=0.01)
                 texture_size = gr.Slider(512, 2048, label="Texture Size", value=1024, step=512)
-            
             with gr.Row():
                 extract_glb_btn = gr.Button("Extract GLB", interactive=False)
                 extract_gs_btn = gr.Button("Extract Gaussian", interactive=False)
-
         with gr.Column():
             video_output = gr.Video(label="Generated 3D Asset", autoplay=True, loop=True, height=300)
-            image_prompt_processed = ImageSlider(label="processed sketch", interactive=False, type="pil", height=512)
+            image_prompt_processed = gr.Image(label="processed sketch", interactive=False, type="pil", height=512)
             model_output = LitModel3D(label="Extracted GLB/Gaussian", exposure=10.0, height=300)
-            
             with gr.Row():
                 download_glb = gr.DownloadButton(label="Download GLB", interactive=False)
                 download_gs = gr.DownloadButton(label="Download Gaussian", interactive=False)  
-    
-    is_multiimage = gr.State(False)
+
     do_preprocess = gr.State(True)
     output_buf = gr.State()
 
-    with gr.Row(visible=False) as single_image_example:
-        examples = gr.Examples(
-            examples=[
-                f'assets/example_image/{image}'
-                for image in os.listdir("assets/example_image")
-            ],
-            inputs=[image_prompt],
-            fn=preprocess_image,
-            outputs=[image_prompt_processed],
-            run_on_click=True,
-            examples_per_page=64,
-        )
-    with gr.Row(visible=False) as multiimage_example:
-        examples_multi = gr.Examples(
-            examples=prepare_multi_example(),
-            inputs=[image_prompt],
-            fn=split_image,
-            outputs=[multiimage_prompt],
-            run_on_click=True,
-            examples_per_page=8,
-        )
-
     demo.load(start_session)
     demo.unload(end_session)
-    
-    multiimage_input_tab.select(
-        lambda: tuple([True, gr.Row.update(visible=False), gr.Row.update(visible=True)]),
-        outputs=[is_multiimage, single_image_example, multiimage_example]
-    )
+
     image_prompt.clear(
         fn=reset_canvas,
         outputs = [image_prompt]
     )
-    
+
     sketch_btn.click(
         get_seed,
         inputs=[randomize_seed, seed],
@@ -393,11 +274,6 @@ with gr.Blocks(delete_cache=(600, 600)) as demo:
         inputs=[image_prompt, prompt, negative_prompt, style, num_steps, guidance_scale, controlnet_conditioning_scale],
         outputs=[image_prompt_processed],
     )
-    multiimage_prompt.upload(
-        preprocess_images,
-        inputs=[multiimage_prompt],
-        outputs=[multiimage_prompt],
-    )
 
     generate_btn.click(
         get_seed,
@@ -405,7 +281,7 @@ with gr.Blocks(delete_cache=(600, 600)) as demo:
         outputs=[seed],
     ).then(
         image_to_3d,
-        inputs=[image_prompt_processed, multiimage_prompt, is_multiimage, seed, ss_guidance_strength, ss_sampling_steps, slat_guidance_strength, slat_sampling_steps, multiimage_algo],
+        inputs=[image_prompt_processed, seed, ss_guidance_strength, ss_sampling_steps, slat_guidance_strength, slat_sampling_steps],
         outputs=[output_buf, video_output],
     ).then(
         lambda: tuple([gr.Button(interactive=True), gr.Button(interactive=True)]),
@@ -425,7 +301,7 @@ with gr.Blocks(delete_cache=(600, 600)) as demo:
         lambda: gr.Button(interactive=True),
         outputs=[download_glb],
     )
-    
+
     extract_gs_btn.click(
         extract_gaussian,
         inputs=[output_buf],
@@ -443,16 +319,13 @@ with gr.Blocks(delete_cache=(600, 600)) as demo:
 if __name__ == "__main__":
     pipeline = TrellisImageTo3DPipeline.from_pretrained("JeffreyXiang/TRELLIS-image-large")
     pipeline.cuda()
-
     device = "cuda" if torch.cuda.is_available() else "cpu"
-    
     #scribble controlnet
     controlnet = ControlNetModel.from_pretrained(
     "xinsir/controlnet-scribble-sdxl-1.0",
     torch_dtype=torch.float16
 )
     vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16)
-    
     pipe_control = StableDiffusionXLControlNetPipeline.from_pretrained(
         "sd-community/sdxl-flash",
         controlnet=controlnet,
@@ -461,7 +334,6 @@ if __name__ == "__main__":
     )
     pipe_control.scheduler = EulerAncestralDiscreteScheduler.from_config(pipe_control.scheduler.config)
     pipe_control.to(device)
-    
     try:
         pipeline.preprocess_image(Image.fromarray(np.zeros((512, 512, 3), dtype=np.uint8)))    # Preload rembg
     except: