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

MODEL_ID = "nvidia/segformer-b5-finetuned-ade-640-640"
processor = AutoImageProcessor.from_pretrained("nvidia/segformer-b0-finetuned-cityscapes-512-1024")
model = AutoModelForSemanticSegmentation.from_pretrained("nvidia/segformer-b0-finetuned-cityscapes-512-1024")

def ade_palette():
    """ADE20K palette that maps each class to RGB values."""
    return [
        [204, 87, 92],[112, 185, 212],[45, 189, 106],[234, 123, 67],[78, 56, 123],[210, 32, 89],
        [90, 180, 56],[155, 102, 200],[33, 147, 176],[255, 183, 76],[67, 123, 89],[190, 60, 45],
        [134, 112, 200],[56, 45, 189],[200, 56, 123],[87, 92, 204],[120, 56, 123],[45, 78, 123],[255, 0, 0],
    ]

labels_list = []
with open("labels.txt", "r", encoding="utf-8") as fp:
    for line in fp:
        labels_list.append(line.rstrip("\n"))

colormap = np.asarray(ade_palette(), dtype=np.uint8)

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]

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=25)
    return fig

def run_inference(input_img):
    # input: numpy array from gradio -> PIL
    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  # (1, C, h/4, w/4)

    # resize to original
    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)  # (H,W)

    # colorize & overlay
    color_seg = colormap[seg]                                # (H,W,3)
    pred_img = (np.array(img) * 0.5 + color_seg * 0.5).astype(np.uint8)

    fig = draw_plot(pred_img, seg)
    return fig

with gr.Blocks(title="City Segmentation Demo") as demo:
    gr.Markdown("# 🏙️ 도시 이미지 시맨틱 세그멘테이션")
    gr.Markdown("이미지를 업로드하면 SegFormer가 도로, 건물, 하늘 등을 색상으로 구분합니다.")

    with gr.Row():
        with gr.Column(scale=1):
            input_image = gr.Image(type="numpy", label="입력 이미지")
            run_button = gr.Button("🔍 분석 실행")
            gr.Examples(
                examples=[
                    ["city-1.jpg"],
                    ["city-2.jpg"],
                    ["city-3.jpeg"],
                    ["city-4.jpg"],
                    ["city-5.jpg"],
                ],
                inputs=input_image
            )

        with gr.Column(scale=2):
            output_plot = gr.Plot(label="결과 (Segmentation Overlay + Legend)")

    run_button.click(fn=run_inference, inputs=input_image, outputs=output_plot)

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