Update app.py

app.py

@@ -1,39 +1,131 @@
 import gradio as gr
+
+from matplotlib import gridspec
+import matplotlib.pyplot as plt
 import numpy as np
 from PIL import Image
 import tensorflow as tf
 from transformers import SegformerFeatureExtractor, TFSegformerForSemanticSegmentation
 
-# Segformer 모델과 관련 객체를 초기화
-feature_extractor = SegformerFeatureExtractor.from_pretrained("nickmuchi/segformer-b4-finetuned-segments-sidewalk")
-model = TFSegformerForSemanticSegmentation.from_pretrained("nickmuchi/segformer-b4-finetuned-segments-sidewalk", from_pt=True)
+feature_extractor = SegformerFeatureExtractor.from_pretrained(
+    "nickmuchi/segformer-b4-finetuned-segments-sidewalk"
+)
+model = TFSegformerForSemanticSegmentation.from_pretrained(
+    "nickmuchi/segformer-b4-finetuned-segments-sidewalk",
+    from_pt=True
+
+)
+
+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],
+        [156, 200, 56],
+        [32, 90, 210],
+        [56, 123, 67],
+        [180, 56, 123],
+        [123, 67, 45],
+        [45, 134, 200],
+        [67, 56, 123],
+        [78, 123, 67],
+        [32, 210, 90],
+        [45, 56, 189],
+        [123, 56, 123],
+        [56, 156, 200],
+        [189, 56, 45],
+        [112, 200, 56],
+        [56, 123, 45],
+        [200, 32, 90],
+        [255, 255, 0],
+    ]
+
+labels_list = []
+
+with open(r'labels.txt', 'r') as fp:
+    for line in fp:
+        labels_list.append(line[:-1])
+
+colormap = np.asarray(ade_palette())
+
+
+def label_to_color_image(label):
+    if label.ndim != 2:
+        raise ValueError("Expect 2-D input label")
 
-def perform_semantic_segmentation(input_img):
+    if np.max(label) >= len(colormap):
+        raise ValueError("label value too large.")
+    return colormap[label]
+
+def draw_plot(pred_img, seg):
+    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.numpy().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 sepia(input_img):
     input_img = Image.fromarray(input_img)
 
-    # 이미지를 처리하고 모델에 전달
     inputs = feature_extractor(images=input_img, return_tensors="tf")
     outputs = model(**inputs)
     logits = outputs.logits
 
-    # 모델 출력을 처리하여 시맨틱 분할 결과를 얻음
     logits = tf.transpose(logits, [0, 2, 3, 1])
-    logits = tf.image.resize(logits, input_img.size[::-1])
+    logits = tf.image.resize(
+        logits, input_img.size[::-1]
+    )  # We reverse the shape of `image` because `image.size` returns width and height.
     seg = tf.math.argmax(logits, axis=-1)[0]
 
-    return input_img, seg.numpy()
+    color_seg = np.zeros(
+        (seg.shape[0], seg.shape[1], 3), dtype=np.uint8
+    )  # height, width, 3
+    for label, color in enumerate(colormap):
+        color_seg[seg.numpy() == label, :] = color
 
-def segformer_interface(input_image):
-    original_image, segmentation_map = perform_semantic_segmentation(input_image)
-    return original_image, segmentation_map
+    # Show image + mask
+    pred_img = np.array(input_img) * 0.5 + color_seg * 0.5
+    pred_img = pred_img.astype(np.uint8)
+
+    fig = draw_plot(pred_img, seg)
+    return fig
+
+demo = gr.Interface(fn=sepia,
+                    inputs=gr.Image(shape=(400, 600)),
+                    outputs=['plot'],
+                    examples=["side-1.jpg", "side-2.jpg", "side-3.jpg", "side-4.jpg", "side-5.jpg", "side-6.jpg"],
+                    allow_flagging='never')
 
-# Gradio 데모 구성
-demo = gr.Interface(
-    fn=segformer_interface,
-    inputs=gr.Image(shape=(400, 600)),
-    outputs=[gr.Image(type="plot"), gr.Image(type="plot")],  # 원본 이미지 및 시맨틱 분할 맵을 출력
-    examples=["side-1.jpg", "side-2.jpg", "side-3.jpg"],
-    allow_flagging='never'
-)
 
-demo.launch()
+demo.launch()