app.py

import gradio as gr
import spaces
from gradio_litmodel3d import LitModel3D

import os
import shutil
os.environ['SPCONV_ALGO'] = 'native'
from typing import *
import torch
import numpy as np
import imageio
import uuid
from easydict import EasyDict as edict
from PIL import Image
from trellis.pipelines import TrellisImageTo3DPipeline
from trellis.representations import Gaussian, MeshExtractResult
from trellis.utils import render_utils, postprocessing_utils


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)


def start_session(req: gr.Request):
    user_dir = os.path.join(TMP_DIR, str(req.session_hash))
    print(f'Creating user directory: {user_dir}')
    os.makedirs(user_dir, exist_ok=True)
    
def end_session(req: gr.Request):
    user_dir = os.path.join(TMP_DIR, str(req.session_hash))
    print(f'Removing user directory: {user_dir}')
    shutil.rmtree(user_dir)

def preprocess_image(image: Image.Image) -> Tuple[str, Image.Image]:
    """
    Preprocess the input image.

    Args:
        image (Image.Image): The input image.

    Returns:
        str: uuid of the trial.
        Image.Image: The preprocessed image.
    """
    processed_image = pipeline.preprocess_image(image)
    return processed_image

def pack_state(gs: Gaussian, mesh: MeshExtractResult, trial_id: str) -> dict:
    return {
        'gaussian': {
            **gs.init_params,
            '_xyz': gs._xyz.cpu().numpy(),
            '_features_dc': gs._features_dc.cpu().numpy(),
            '_scaling': gs._scaling.cpu().numpy(),
            '_rotation': gs._rotation.cpu().numpy(),
            '_opacity': gs._opacity.cpu().numpy(),
        },
        'mesh': {
            'vertices': mesh.vertices.cpu().numpy(),
            'faces': mesh.faces.cpu().numpy(),
        },
        'trial_id': trial_id,
    }
    
def unpack_state(state: dict) -> Tuple[Gaussian, edict, str]:
    gs = Gaussian(
        aabb=state['gaussian']['aabb'],
        sh_degree=state['gaussian']['sh_degree'],
        mininum_kernel_size=state['gaussian']['mininum_kernel_size'],
        scaling_bias=state['gaussian']['scaling_bias'],
        opacity_bias=state['gaussian']['opacity_bias'],
        scaling_activation=state['gaussian']['scaling_activation'],
    )
    gs._xyz = torch.tensor(state['gaussian']['_xyz'], device='cuda')
    gs._features_dc = torch.tensor(state['gaussian']['_features_dc'], device='cuda')
    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, state['trial_id']

def get_seed(randomize_seed: bool, seed: int) -> int:
    """
    Get the random seed.
    """
    return np.random.randint(0, MAX_SEED) if randomize_seed else seed

@spaces.GPU
def image_to_3d(
    image: Image.Image,
    seed: int,
    ss_guidance_strength: float,
    ss_sampling_steps: int,
    slat_guidance_strength: float,
    slat_sampling_steps: int,
    req: gr.Request,
) -> Tuple[dict, str]:
    """
    Convert an image to a 3D model with memory management.
    """
    user_dir = os.path.join(TMP_DIR, str(req.session_hash))
    
    # Generate base outputs
    outputs = pipeline.run(
        image,
        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,
        },
    )

    # Clear CUDA cache after model generation
    torch.cuda.empty_cache()
    
    # Generate video preview in smaller batches
    video = []
    video_geo = []
    batch_size = 30  # Process 30 frames at a time
    num_frames = 120
    
    for i in range(0, num_frames, batch_size):
        end_idx = min(i + batch_size, num_frames)
        curr_frames = end_idx - i
        
        # Generate color frames
        batch_frames = render_utils.render_video(
            outputs['gaussian'][0],
            num_frames=curr_frames,
            start_frame=i
        )['color']
        video.extend(batch_frames)
        
        # Generate geometry frames
        batch_geo = render_utils.render_video(
            outputs['mesh'][0],
            num_frames=curr_frames,
            start_frame=i
        )['normal']
        video_geo.extend(batch_geo)
        
        # Clear cache after each batch
        torch.cuda.empty_cache()

    # Combine and save video
    video = [np.concatenate([video[i], video_geo[i]], axis=1) for i in range(len(video))]
    trial_id = str(uuid.uuid4())
    video_path = os.path.join(user_dir, f"{trial_id}.mp4")
    imageio.mimsave(video_path, video, fps=15)
    
    # Clear memory
    del video
    del video_geo
    torch.cuda.empty_cache()
    
    # Pack state and return
    state = pack_state(outputs['gaussian'][0], outputs['mesh'][0], trial_id)
    return state, video_path

@spaces.GPU
def export_full_quality_glb(
    state: dict,
    req: gr.Request,
) -> Tuple[str, str]:
    """
    Export a full-quality GLB file with memory management.
    """
    user_dir = os.path.join(TMP_DIR, str(req.session_hash))
    gs, mesh, trial_id = unpack_state(state)
    
    # Clear cache before starting
    torch.cuda.empty_cache()
    
    glb = postprocessing_utils.to_glb(
        gs,
        mesh,
        simplify=0.0,  # No simplification
        fill_holes=True,
        fill_holes_max_size=0.04,
        texture_size=2048,  # Maximum texture resolution
        verbose=True  # Show progress
    )
    glb_path = os.path.join(user_dir, f"{trial_id}_full.glb")
    glb.export(glb_path)
    
    # Clear cache after finishing
    torch.cuda.empty_cache()
    return glb_path, glb_path

@spaces.GPU
def extract_glb(
    state: dict,
    mesh_simplify: float,
    texture_size: int,
    req: gr.Request,
) -> Tuple[str, str]:
    """
    Extract a GLB file from the 3D model.
    """
    user_dir = os.path.join(TMP_DIR, str(req.session_hash))
    gs, mesh, trial_id = unpack_state(state)
    glb = postprocessing_utils.to_glb(gs, mesh, simplify=mesh_simplify, texture_size=texture_size, verbose=False)
    glb_path = os.path.join(user_dir, f"{trial_id}.glb")
    glb.export(glb_path)
    return glb_path, glb_path

with gr.Blocks(delete_cache=(600, 600)) as demo:
    gr.Markdown("""
    ## Image to 3D Asset with [TRELLIS](https://trellis3d.github.io/)
    * Upload an image and click "Generate" to create a 3D asset. If the image has alpha channel, it be used as the mask. Otherwise, we use `rembg` to remove the background.
    * After generation:
        * Click "Download Full-Quality GLB" for maximum quality
        * Or use GLB Extraction Settings for a reduced size version
    """)
    
    with gr.Row():
        with gr.Column():
            image_prompt = gr.Image(label="Image Prompt", format="png", image_mode="RGBA", type="pil", height=300)
            
            with gr.Accordion(label="Generation Settings", open=False):
                seed = gr.Slider(0, MAX_SEED, label="Seed", value=0, step=1)
                randomize_seed = gr.Checkbox(label="Randomize Seed", value=True)
                gr.Markdown("Stage 1: Sparse Structure Generation")
                with gr.Row():
                    ss_guidance_strength = gr.Slider(0.0, 10.0, label="Guidance Strength", value=7.5, step=0.1)
                    ss_sampling_steps = gr.Slider(1, 500, label="Sampling Steps", value=12, step=1)
                gr.Markdown("Stage 2: Structured Latent Generation")
                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, 500, label="Sampling Steps", value=12, step=1)

            generate_btn = gr.Button("Generate")
            
            with gr.Accordion(label="GLB Extraction Settings", open=False):
                mesh_simplify = gr.Slider(0.0, 0.98, label="Simplify", value=0.95, step=0.01)
                texture_size = gr.Slider(512, 2048, label="Texture Size", value=1024, step=512)
            
            extract_glb_btn = gr.Button("Extract GLB", interactive=False)

        with gr.Column():
            video_output = gr.Video(label="Generated 3D Asset", autoplay=True, loop=True, height=300)
            model_output = LitModel3D(label="3D Model Preview", exposure=20.0, height=300)
            with gr.Row():
                download_full = gr.DownloadButton(label="Download Full-Quality GLB", interactive=False)
                download_reduced = gr.DownloadButton(label="Download Reduced GLB", interactive=False)
            
    output_buf = gr.State()

    # Example images at the bottom of the page
    with gr.Row():
        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],
            run_on_click=True,
            examples_per_page=64,
        )

    # Event handlers
    demo.load(start_session)
    demo.unload(end_session)
    
    image_prompt.upload(
        preprocess_image,
        inputs=[image_prompt],
        outputs=[image_prompt],
    )

    generate_btn.click(
        get_seed,
        inputs=[randomize_seed, seed],
        outputs=[seed],
    ).then(
        image_to_3d,
        inputs=[image_prompt, seed, ss_guidance_strength, ss_sampling_steps, slat_guidance_strength, slat_sampling_steps],
        outputs=[output_buf, video_output],
    ).then(
        lambda: [gr.Button(interactive=True), gr.Button(interactive=True), gr.Button(interactive=False)],
        outputs=[download_full, extract_glb_btn, download_reduced],
    )

    download_full.click(
        export_full_quality_glb,
        inputs=[output_buf],
        outputs=[model_output, download_full],
    ).then(
        lambda: gr.Button(interactive=True),
        outputs=[download_full],
    )

    extract_glb_btn.click(
        extract_glb,
        inputs=[output_buf, mesh_simplify, texture_size],
        outputs=[model_output, download_reduced],
    ).then(
        lambda: gr.Button(interactive=True),
        outputs=[download_reduced],
    )

if __name__ == "__main__":
    pipeline = TrellisImageTo3DPipeline.from_pretrained("JeffreyXiang/TRELLIS-image-large")
    pipeline.cuda()
    try:
        pipeline.preprocess_image(Image.fromarray(np.zeros((512, 512, 3), dtype=np.uint8)))
    except:
        pass
    demo.launch()