1

app.py

@@ -1,111 +1,36 @@
 import gradio as gr
-
-from matplotlib import gridspec
-import matplotlib.pyplot as plt
-import numpy as np
+from transformers import SegformerFeatureExtractor, SegformerForSemanticSegmentation
 from PIL import Image
-import tensorflow as tf
-from transformers import SegformerFeatureExtractor, TFSegformerForSemanticSegmentation
-
-feature_extractor = SegformerFeatureExtractor.from_pretrained(
-    "nvidia/segformer-b5-finetuned-cityscapes-1024-1024"
-)
-model = TFSegformerForSemanticSegmentation.from_pretrained(
-    "nvidia/segformer-b5-finetuned-cityscapes-1024-1024"
-)
-
-def ade_palette():
-    """ADE20K palette that maps each class to RGB values."""
-    return [
-        [255,0,0], #빨강
-        [255,228,0], #노랑
-        [171,242,0], # 연두
-        [0,216,255], #하늘
-        [0,0,255], #파랑
-        [255,0,221], #핑크
-        [116,116,116], #회색
-        [95,0,255], #보라
-        [255,94,0], #주황
-        [71,200,62], #초록
-        [153,0,76], #마젠타
-        [67,116,217], #애매한하늘 + 파랑
-        [153,112,0], #겨자
-        [87,129,0], #녹색
-        [255,169,169], #분홍분홍
-        [35,30,183], #어두운 파랑
-        [225,186,133], #살색
-        [206,251,201], #연한초록
-        [165,102,255] #애매한 보라
-    ]
-
-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")
-
-    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)
+import numpy as np
 
-    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)
+# 모델과 특징 추출기 불러오기
+feature_extractor = SegformerFeatureExtractor.from_pretrained("nvidia/segformer-b5-finetuned-cityscapes-1024-1024")
+model = SegformerForSemanticSegmentation.from_pretrained("nvidia/segformer-b5-finetuned-cityscapes-1024-1024")
 
-    inputs = feature_extractor(images=input_img, return_tensors="tf")
+def segment_image(image):
+    # 이미지를 처리하고 모델에 전달하기
+    inputs = feature_extractor(images=image, return_tensors="pt")
     outputs = model(**inputs)
     logits = outputs.logits
 
-    logits = tf.transpose(logits, [0, 2, 3, 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]
-
-    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
-
-    # 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=["image1", "image2", "image3"],
-                    allow_flagging='never')
-
+    # 결과 처리 및 이미지로 변환
+    result = logits.argmax(dim=1)[0]
+    result = result.cpu().detach().numpy()
+    result_image = Image.fromarray(result.astype(np.uint8), mode="P")
+
+    # 결과 이미지 반환
+    return result_image
+
+# Gradio 인터페이스 정의
+iface = gr.Interface(
+    fn=segment_image,
+    inputs=gr.inputs.Image(type="pil"),
+    examples = ['image1.jpg', 'image2.jpg', 'image3.jpg'],
+    outputs=['plot'],
+    title="SegFormer Image Segmentation",
+    description="Upload an image to segment it using the SegFormer model trained on Cityscapes dataset."
+)
 
-demo.launch()
+# 인터페이스 실행
+iface.launch()