| """ |
| Visualize sample images and masks from the CMP Facade dataset. |
| """ |
| import os |
| import numpy as np |
| import matplotlib |
| matplotlib.use('Agg') |
| import matplotlib.pyplot as plt |
| from PIL import Image |
| import random |
|
|
|
|
| def visualize_samples(data_dir="data/cmp_facade", split="train", num_samples=5, save_path="outputs/dataset_visualization.png"): |
| os.makedirs(os.path.dirname(save_path), exist_ok=True) |
|
|
| split_dir = f"{data_dir}/{split}" |
| img_dir = f"{split_dir}/images" |
| mask_dir = f"{split_dir}/masks_binary" |
|
|
| img_files = sorted([f for f in os.listdir(img_dir) if f.endswith('.jpg')])[:20] |
| selected = random.sample(img_files, min(num_samples, len(img_files))) |
|
|
| fig, axes = plt.subplots(num_samples, 3, figsize=(12, 4 * num_samples)) |
| if num_samples == 1: |
| axes = axes.reshape(1, -1) |
|
|
| for i, img_file in enumerate(selected): |
| img = Image.open(f"{img_dir}/{img_file}").convert('RGB') |
| mask = Image.open(f"{mask_dir}/{img_file.replace('.jpg', '.png')}").convert('L') |
| mask_arr = np.array(mask) |
|
|
| img_arr = np.array(img) |
| overlay = img_arr.copy() |
| overlay[mask_arr > 0] = [255, 0, 0] |
| blended = (img_arr * 0.6 + overlay * 0.4).astype(np.uint8) |
|
|
| axes[i, 0].imshow(img) |
| axes[i, 0].set_title(f"Image {img_file}") |
| axes[i, 0].axis('off') |
|
|
| axes[i, 1].imshow(mask, cmap='gray') |
| axes[i, 1].set_title("Facade Mask (class 1)") |
| axes[i, 1].axis('off') |
|
|
| axes[i, 2].imshow(blended) |
| axes[i, 2].set_title("Overlay") |
| axes[i, 2].axis('off') |
|
|
| plt.tight_layout() |
| plt.savefig(save_path, dpi=150) |
| plt.close() |
| print(f"Visualization saved to {save_path}") |
|
|
|
|
| if __name__ == "__main__": |
| visualize_samples("data/cmp_facade", "train", num_samples=5, save_path="outputs/dataset_visualization.png") |
|
|
