app.py

import gradio as gr
from matplotlib import gridspec
import matplotlib.pyplot as plt
import numpy as np
from PIL import Image
import torch
from transformers import AutoImageProcessor, AutoModelForSemanticSegmentation

# ==============================
# ✅ 모델 로드
# ==============================
MODEL_ID = "nvidia/segformer-b2-finetuned-cityscapes-1024-1024"
processor = AutoImageProcessor.from_pretrained(MODEL_ID)
model = AutoModelForSemanticSegmentation.from_pretrained(MODEL_ID)

# ==============================
# ✅ 팔레트 / 라벨 로드
# ==============================
def city_palette():
    return [
        [128, 64, 128], [244, 35, 232], [70, 70, 70], [102, 102, 156], [190, 153, 153],
        [153, 153, 153], [250, 170, 30], [220, 220, 0], [107, 142, 35], [152, 251, 152],
        [70, 130, 180], [220, 20, 60], [255, 0, 0], [0, 0, 142], [0, 0, 70],
        [0, 60, 100], [0, 80, 100], [0, 0, 230], [119, 11, 32]
    ]

colormap = np.asarray(city_palette(), dtype=np.uint8)
labels_list = [l.strip() for l in open("labels.txt", "r", encoding="utf-8").readlines()]

def label_to_color_image(label):
    if label.ndim != 2:
        raise ValueError("Expect 2-D input label")
    if np.max(label) >= len(colormap):
        raise ValueError("Label value too large.")
    return colormap[label]

# ==============================
# ✅ 시각화 함수
# ==============================
def draw_plot(pred_img, seg_np):
    fig = plt.figure(figsize=(20, 15))
    grid_spec = gridspec.GridSpec(1, 2, width_ratios=[6, 1])

    plt.subplot(grid_spec[0])
    plt.imshow(pred_img)
    plt.axis('off')

    LABEL_NAMES = np.asarray(labels_list)
    FULL_LABEL_MAP = np.arange(len(LABEL_NAMES)).reshape(len(LABEL_NAMES), 1)
    FULL_COLOR_MAP = label_to_color_image(FULL_LABEL_MAP)

    unique_labels = np.unique(seg_np.astype("uint8"))
    ax = plt.subplot(grid_spec[1])
    plt.imshow(FULL_COLOR_MAP[unique_labels].astype(np.uint8), interpolation="nearest")
    ax.yaxis.tick_right()
    plt.yticks(range(len(unique_labels)), LABEL_NAMES[unique_labels])
    plt.xticks([], [])
    ax.tick_params(width=0.0, labelsize=22)
    plt.tight_layout()
    return fig

# ==============================
# ✅ 추론 함수
# ==============================
def run_inference(input_img):
    img = Image.fromarray(input_img.astype(np.uint8)) if isinstance(input_img, np.ndarray) else input_img
    if img.mode != "RGB":
        img = img.convert("RGB")

    inputs = processor(images=img, return_tensors="pt")
    with torch.no_grad():
        outputs = model(**inputs)
        logits = outputs.logits

    upsampled = torch.nn.functional.interpolate(
        logits, size=img.size[::-1], mode="bilinear", align_corners=False
    )
    seg = upsampled.argmax(dim=1)[0].cpu().numpy().astype(np.uint8)
    color_seg = colormap[seg]
    pred_img = (np.array(img) * 0.5 + color_seg * 0.5).astype(np.uint8)
    return draw_plot(pred_img, seg)

# ==============================
# 🌆 Gradio UI 구성
# ==============================
with gr.Blocks(theme=gr.themes.Soft(), css="""
#title {text-align:center; font-size:2.2em; font-weight:700; margin-bottom:0.5em;}
#desc {text-align:center; color:#555; font-size:1.1em; margin-bottom:2em;}
#footer {text-align:center; color:#888; margin-top:2em; font-size:0.9em;}
""") as demo:
    
    gr.Markdown("<div id='title'>🏙️ City Segmenter</div>")
    gr.Markdown("<div id='desc'>SegFormer-B2 기반 도시 이미지 세그멘테이션.<br>도로, 건물, 차량, 사람 등을 색상으로 분리하여 시각화합니다.</div>")

    with gr.Row():
        with gr.Column(scale=1):
            input_img = gr.Image(type="numpy", label="📤 이미지 업로드")
            example_box = gr.Examples(
                examples=[
                    "city1.jpeg",
                    "city2.jpg",
                    "city3.jpg"
                ],
                inputs=input_img,
                label="🖼️ 예시 이미지"
            )
            run_btn = gr.Button("🚀 분석 시작", variant="primary")
        with gr.Column(scale=2):
            output_plot = gr.Plot(label="🧭 결과 시각화 (Overlay + Legend)")

    run_btn.click(fn=run_inference, inputs=input_img, outputs=output_plot)

    gr.Markdown("<div id='footer'>© 2025 City Segmenter • Powered by NVIDIA SegFormer-B2</div>")

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