I am trying to deform an image using the following vector field. I have tried to use the response in this post. However, all of my attempts have been unsuccessful till now. The deformation field is generated using the following code: import numpy as np import cv2 size = 100 x = np.linspace(-1, 1, size) y = np.linspace(-1, 1, size) X, Y = np.meshgrid(x, y) phase_profile = 800.0*((X-0.0)**2 + (Y-0.0)**2) How to deform an image given the deformation field represented by the light blue arrows ? Here is an example of an image I am trying to deform. My attempt is the following: def stack_overflow_gradient_distortion_map(phase_profile, size=256, coeff=30.0): # Create coordinate grids x = np.linspace(-1, 1, size) y = np.linspace(-1, 1, size) X, Y = np.meshgrid(x, y) shape = X.shape dx = 2 / (size - 1) grad_y, grad_x = np.gradient(phase_profile, dx, dx) # Compute gradients (deformation field) gy, gx = np.gradient(phase_profile) mapx_base, mapy_base = np.meshgrid(np.arange(shape[0]), np.arange(shape[1])) mapx = mapx_base + gx*coeff mapy = mapy_base + gy*coeff return mapx, mapy P.S: The plot function used is the following: def plot_phase_heatmap_with_gradients(phase_profile, mesh_grid_x, mesh_grid_y): dx = 2 / (size - 1) # Physical spacing in the [-1, 1] range grad_y, grad_x = np.gradient(phase_profile, dx, dx) plt.figure(figsize=(8, 6)) plt.imshow(phase_profile, cmap='hot', origin='lower', extent=[-1, 1, -1, 1]) plt.colorbar(label='Phase Profile') print("max grad_x: ", np.max(grad_x)) # Overlay gradient arrows skip = 10 # Adjust to reduce arrow density plt.quiver(mesh_grid_x[::skip, ::skip], mesh_grid_y[::skip, ::skip], grad_x[::skip, ::skip], grad_y[::skip, ::skip], color='cyan') plt.xlabel('X-axis') plt.ylabel('Y-axis') plt.title('Phase Profile with Gradient Vectors') plt.show() Continue reading...
