Open Frreed opened 1 year ago
This is how I preprocess the dataset
near_points = [
surface_points + np.random.normal(scale=0.005, size=(N_near, 3)),
surface_points + np.random.normal(scale=0.05, size=(N_near, 3)),
]
near_points = np.concatenate(near_points)
Given a surface point cloud ( surface_points
), we jitter the point cloud with Gaussian noise.
@1zb Hi, very thanks for your reply, good luck to you!
Hi, very thanks for your amazing work. i want to ask how to generate near_pionts myself ?