parkchihoon/MachineLearning

1#수정 코드2import gradio as gr3 4from matplotlib import gridspec5import matplotlib.pyplot as plt6import numpy as np7from PIL import Image8import tensorflow as tf9from transformers import SegformerFeatureExtractor, TFSegformerForSemanticSegmentation10 11feature_extractor = SegformerFeatureExtractor.from_pretrained(12    "mattmdjaga/segformer_b2_clothes"13)14model = TFSegformerForSemanticSegmentation.from_pretrained(15    "mattmdjaga/segformer_b2_clothes"16)17 18def ade_palette():19    """ADE20K palette that maps each class to RGB values."""20    return [21        [204, 87, 92],22        [112, 185, 212],23        [45, 189, 106],24        [234, 123, 67],25        [78, 56, 123],26        [210, 32, 89],27        [90, 180, 56],28        [155, 102, 200],29        [33, 147, 176],30        [255, 183, 76],31        [67, 123, 89],32        [190, 60, 45],33        [134, 112, 200],34        [56, 45, 189],35        [200, 56, 123],36        [87, 92, 204],37        [120, 56, 123],38    ]39 40labels_list = []41 42with open(r'labels.txt', 'r') as fp:43    for line in fp:44        labels_list.append(line[:-1])45 46colormap = np.asarray(ade_palette())47 48def label_to_color_image(label):49    if label.ndim != 2:50        raise ValueError("Expect 2-D input label")51 52    if np.max(label) >= len(colormap):53        raise ValueError("label value too large.")54    return colormap[label]55 56def draw_plot(pred_img, seg):57    fig = plt.figure(figsize=(20, 15))58 59    grid_spec = gridspec.GridSpec(1, 2, width_ratios=[6, 1])60 61    plt.subplot(grid_spec[0])62    plt.imshow(pred_img)63    plt.axis('off')64    LABEL_NAMES = np.asarray(labels_list)65    FULL_LABEL_MAP = np.arange(len(LABEL_NAMES)).reshape(len(LABEL_NAMES), 1)66    FULL_COLOR_MAP = label_to_color_image(FULL_LABEL_MAP)67 68    unique_labels = np.unique(seg.numpy().astype("uint8"))69    ax = plt.subplot(grid_spec[1])70    plt.imshow(FULL_COLOR_MAP[unique_labels].astype(np.uint8), interpolation="nearest")71    ax.yaxis.tick_right()72    plt.yticks(range(len(unique_labels)), LABEL_NAMES[unique_labels])73    plt.xticks([], [])74    ax.tick_params(width=0.0, labelsize=25)75    return fig76 77def sepia(input_img):78    input_img = Image.fromarray(input_img)79 80    inputs = feature_extractor(images=input_img, return_tensors="tf")81    outputs = model(**inputs)82    logits = outputs.logits83 84    logits = tf.transpose(logits, [0, 2, 3, 1])85    logits = tf.image.resize(86        logits, input_img.size[::-1]87    )  # We reverse the shape of `image` because `image.size` returns width and height.88    seg = tf.math.argmax(logits, axis=-1)[0]89 90    color_seg = np.zeros(91        (seg.shape[0], seg.shape[1], 3), dtype=np.uint892    )  # height, width, 393    for label, color in enumerate(colormap):94        color_seg[seg.numpy() == label, :] = color95 96    # Show image + mask97    pred_img = np.array(input_img) * 0.5 + color_seg * 0.598    pred_img = pred_img.astype(np.uint8)99 100    fig = draw_plot(pred_img, seg)101    return fig102 103demo = gr.Interface(fn=sepia,104                    inputs=gr.Image(shape=[400, 600]),105                    outputs=['plot'],106                    examples=['person-1.jpg', 'person-2.jpg', 'person-3.jpg', 'person-4.jpg', 'person-5.jpg'],107                    allow_flagging='never')108 109 110demo.launch()111