app.py

import gradio as gr
import torch
import numpy as np
from PIL import Image
from torchvision.models.segmentation import deeplabv3_resnet50, DeepLabV3_ResNet50_Weights
from huggingface_hub import hf_hub_download
import cv2

# ---------------- 下载并加载 LaMa 官方权重 ----------------
repo_id = "JosephCatrambone/big-lama-torchscript"
model_path = hf_hub_download(repo_id=repo_id, filename="lama.pt")
lama_model = torch.jit.load(model_path, map_location="cpu")
lama_model.eval()

# ---- 加载分割模型（CPU） ----
device = torch.device("cpu")
weights = DeepLabV3_ResNet50_Weights.COCO_WITH_VOC_LABELS_V1
model = deeplabv3_resnet50(weights=weights).to(device).eval()
preprocess = weights.transforms()

MAX_SIDE = 1024  # 为了速度与内存，限制输入最大边

def _resize_if_needed(pil_img: Image.Image, max_side=MAX_SIDE) -> Image.Image:
    w, h = pil_img.size
    if max(w, h) <= max_side:
        return pil_img
    r = max_side / float(max(w, h))
    return pil_img.resize((int(w * r), int(h * r)), Image.BILINEAR)

def segment(image: Image.Image):
    if not isinstance(image, Image.Image):
        image = Image.fromarray(image)

    image = image.convert("RGB")
    image = _resize_if_needed(image)

    # 预处理并推理
    x = torch.from_numpy(np.array(image)).permute(2, 0, 1).float() / 255.0
    x = x.unsqueeze(0).to(device)  # [1,3,H,W]

    with torch.no_grad():
        out = model(x)["out"][0]                   # [C,H,W]，C=21
    pred = out.argmax(0).cpu().numpy()             # [H,W]

    # 前景 = 非背景（背景类在COCO VOC权重下是0）
    fg = (pred != 0).astype(np.uint8)

    # ---------------- mask ----------------
    kernel = np.ones((15,15), np.uint8)
    fg_dilated = cv2.dilate(fg, kernel, iterations=1)

    mask_img = Image.fromarray((fg_dilated * 255).astype(np.uint8), mode="L")

    # 叠加彩色遮罩（红色半透明）
    base = image.convert("RGBA")
    overlay = Image.new("RGBA", base.size, (255, 0, 0, 0))
    alpha = Image.fromarray((fg_dilated * 120).astype(np.uint8))
    overlay.putalpha(alpha)
    blended = Image.alpha_composite(base, overlay).convert("RGB")

    # ---- LaMa ----
    img_np = np.array(image)  # HWC, uint8
    mask_np = np.array(mask_img)  # H,W, 0/255
    img_t = torch.from_numpy(img_np).permute(2, 0, 1).float().unsqueeze(0) / 255.0
    mask_t = torch.from_numpy(mask_np).unsqueeze(0).unsqueeze(0).float() / 255.0
    with torch.no_grad():
        inpainted_t = lama_model(img_t, mask_t)  # [1,3,H,W]
    inpainted_np = (inpainted_t[0].permute(1, 2, 0).numpy() * 255).astype(np.uint8)

    # ---- 裁剪回原图大小 ----
    H, W = img_np.shape[:2]
    inpainted_np = inpainted_np[:H, :W, :]
    inpainted_img = Image.fromarray(inpainted_np)

    return blended, mask_img, inpainted_img

# ---- Gradio 界面 ----
demo = gr.Interface(
    fn=segment,
    inputs=gr.Image(type="pil", label="Upload Image"),
    outputs=[
        gr.Image(type="pil", label="Overlay (foreground)"),
        gr.Image(type="pil", label="Binary Mask (foreground=white)"),
        gr.Image(type="pil", label="inpaint result"),
    ],
    title="Semantic Segmentation + LaMa Inpainting",
    description="DeepLabV3 segmentation + LaMa inpainting。",
    examples=[
        ["./9F27E2C4-5662-4AA7-A14A-2DE6627EBE8E-14319-000010D528167C0B.PNG"]  
    ]
)

if __name__ == "__main__":
    demo.launch()