173 lines
6.5 KiB
Python
173 lines
6.5 KiB
Python
from pathlib import Path
|
|
import numpy as np
|
|
from PIL import Image, ImageDraw
|
|
import matplotlib.pyplot as plt
|
|
from matplotlib.tri import Triangulation
|
|
|
|
from image_preparing import resize_image_to_resolution
|
|
|
|
import time
|
|
|
|
|
|
transformation_resolution = (1600, 1200)
|
|
|
|
def create_grid_correspondences(image, nx=8, ny=6):
|
|
w, h = image.size
|
|
xs = np.linspace(0, w - 1, nx)
|
|
ys = np.linspace(0, h - 1, ny)
|
|
pts = []
|
|
for y in ys:
|
|
for x in xs:
|
|
pts.append([float(x), float(y)])
|
|
return np.array(pts, dtype=float).copy()
|
|
def calculate_delaunay_triangles(points):
|
|
tri = Triangulation(points[:, 0], points[:, 1])
|
|
return tri.triangles
|
|
def calculate_affine_coefficients(from_triangle, to_triangle):
|
|
A = []
|
|
B = []
|
|
for (x_dst, y_dst), (x_src, y_src) in zip(from_triangle, to_triangle):
|
|
A.append([x_dst, y_dst, 1, 0, 0, 0])
|
|
A.append([0, 0, 0, x_dst, y_dst, 1])
|
|
B.append(x_src)
|
|
B.append(y_src)
|
|
sol, *_ = np.linalg.lstsq(np.array(A), np.array(B), rcond=None)
|
|
return tuple(sol.tolist())
|
|
def warp_triangle(source_image, source_triangle, destination_triangle, output_size, offset):
|
|
dx, dy = offset
|
|
destination_local = np.array(destination_triangle) - np.array([dx, dy])
|
|
coefficients = calculate_affine_coefficients(destination_local, source_triangle)
|
|
return source_image.transform(output_size, Image.AFFINE, coefficients, resample=Image.BILINEAR)
|
|
def reconstruct_goal_from_source(source_image, source_points, destination_points):
|
|
w, h = source_image.size
|
|
triangles = calculate_delaunay_triangles(destination_points)
|
|
out = Image.new("RGB", (w, h))
|
|
for tri in triangles:
|
|
tri = np.asarray(tri, dtype=int)
|
|
source_triangle = source_points[tri]
|
|
destination_triangle = destination_points[tri]
|
|
min_x = max(int(np.floor(destination_triangle[:, 0].min())), 0)
|
|
max_x = min(int(np.ceil(destination_triangle[:, 0].max())), w)
|
|
min_y = max(int(np.floor(destination_triangle[:, 1].min())), 0)
|
|
max_y = min(int(np.ceil(destination_triangle[:, 1].max())), h)
|
|
if max_x <= min_x or max_y <= min_y:
|
|
continue
|
|
out_w = max_x - min_x
|
|
out_h = max_y - min_y
|
|
warped = warp_triangle(source_image, source_triangle, destination_triangle, (out_w, out_h), (min_x, min_y))
|
|
mask = Image.new("L", (out_w, out_h), 0)
|
|
draw = ImageDraw.Draw(mask)
|
|
tri_local = [(x - min_x, y - min_y) for x, y in destination_triangle.tolist()]
|
|
draw.polygon(tri_local, fill=255)
|
|
out.paste(warped, (min_x, min_y), mask)
|
|
return out
|
|
def show_image_transition(source_path, goal_path):
|
|
src = resize_image_to_resolution(source_path, size=transformation_resolution)
|
|
goal = resize_image_to_resolution(goal_path, size=transformation_resolution)
|
|
start = time.time()
|
|
image = reposition_image_tiles(src, goal, grid=transformation_resolution)
|
|
end = time.time()
|
|
print(f"Transition took {end - start:.2f} seconds")
|
|
fig, ax = plt.subplots(figsize=(8, 6), facecolor="black")
|
|
fig.patch.set_facecolor("black")
|
|
ax.set_facecolor("black")
|
|
ax.imshow(image, cmap=None)
|
|
ax.axis("off")
|
|
|
|
def close_transition_on_key(event):
|
|
if event.key in ("q", "escape"):
|
|
plt.close(fig)
|
|
|
|
fig.canvas.mpl_connect("key_press_event", close_transition_on_key)
|
|
|
|
plt.show()
|
|
def calculate_grid_edges(size, divisions):
|
|
return np.rint(np.linspace(0, size, divisions + 1)).astype(np.int32)
|
|
def build_tile_boxes_and_centers(width, height, columns, rows):
|
|
x_edges = calculate_grid_edges(width, columns)
|
|
y_edges = calculate_grid_edges(height, rows)
|
|
|
|
x0 = np.repeat(x_edges[:-1], rows)
|
|
x1 = np.repeat(x_edges[1:], rows)
|
|
y0 = np.tile(y_edges[:-1], columns)
|
|
y1 = np.tile(y_edges[1:], columns)
|
|
|
|
boxes = np.stack([x0, y0, x1, y1], axis=1)
|
|
centers = np.stack([(x0 + x1) * 0.5, (y0 + y1) * 0.5], axis=1).astype(np.float32)
|
|
areas = np.maximum((x1 - x0) * (y1 - y0), 1).astype(np.float32)
|
|
return boxes, centers, areas
|
|
def calculate_integral_image_rgb(array):
|
|
integral = array.cumsum(axis=0).cumsum(axis=1)
|
|
return np.pad(integral, ((1, 0), (1, 0), (0, 0)), mode="constant", constant_values=0)
|
|
def calculate_tile_means(integral, boxes, areas):
|
|
x0 = boxes[:, 0]
|
|
y0 = boxes[:, 1]
|
|
x1 = boxes[:, 2]
|
|
y1 = boxes[:, 3]
|
|
|
|
sums = (
|
|
integral[y1, x1]
|
|
- integral[y0, x1]
|
|
- integral[y1, x0]
|
|
+ integral[y0, x0]
|
|
)
|
|
return sums / areas[:, None]
|
|
def match_tiles_by_feature_rank(source_averages, source_centers, goal_averages, goal_centers, width, height):
|
|
color_scale = 255.0 * np.sqrt(3.0)
|
|
spatial_scale = max(np.hypot(width, height), 1e-6)
|
|
|
|
src_feat = np.hstack([
|
|
source_averages / color_scale,
|
|
(source_centers / spatial_scale) * 0.35,
|
|
])
|
|
goal_feat = np.hstack([
|
|
goal_averages / color_scale,
|
|
(goal_centers / spatial_scale) * 0.35,
|
|
])
|
|
|
|
# Fixed projection keeps pairing deterministic and avoids O(n^2) matching.
|
|
projection = np.array([0.50, 0.30, 0.20, 0.60, 0.40], dtype=np.float32)
|
|
src_order = np.argsort(src_feat @ projection)
|
|
goal_order = np.argsort(goal_feat @ projection)
|
|
return src_order, goal_order
|
|
def reposition_image_tiles(source_image, goal_image, grid=transformation_resolution):
|
|
src = source_image.convert("RGB")
|
|
goal = goal_image.convert("RGB")
|
|
w, h = src.size
|
|
nx, ny = grid
|
|
|
|
src_boxes, src_centers, src_areas = build_tile_boxes_and_centers(w, h, nx, ny)
|
|
goal_boxes, goal_centers, goal_areas = build_tile_boxes_and_centers(w, h, nx, ny)
|
|
|
|
src_arr = np.asarray(src, dtype=np.float32)
|
|
goal_arr = np.asarray(goal, dtype=np.float32)
|
|
src_integral = calculate_integral_image_rgb(src_arr)
|
|
goal_integral = calculate_integral_image_rgb(goal_arr)
|
|
|
|
src_avgs = calculate_tile_means(src_integral, src_boxes, src_areas)
|
|
goal_avgs = calculate_tile_means(goal_integral, goal_boxes, goal_areas)
|
|
|
|
src_order, goal_order = match_tiles_by_feature_rank(
|
|
src_avgs,
|
|
src_centers,
|
|
goal_avgs,
|
|
goal_centers,
|
|
w,
|
|
h,
|
|
)
|
|
|
|
final = Image.new("RGB", (w, h))
|
|
for s_idx, g_idx in zip(src_order.tolist(), goal_order.tolist()):
|
|
sx0, sy0, sx1, sy1 = src_boxes[s_idx].tolist()
|
|
gx0, gy0, gx1, gy1 = goal_boxes[g_idx].tolist()
|
|
|
|
tile = src.crop((sx0, sy0, sx1, sy1))
|
|
goal_w = max(gx1 - gx0, 1)
|
|
goal_h = max(gy1 - gy0, 1)
|
|
if tile.size != (goal_w, goal_h):
|
|
tile = tile.resize((goal_w, goal_h), Image.Resampling.BILINEAR)
|
|
final.paste(tile, (gx0, gy0))
|
|
return final
|
|
|
|
|