ENSEEIHT/projet-probleme-inverse-3D
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity

fomat: apply black formatting

Browse source
This commit is contained in:
Laureηt 2023-01-25 19:54:28 +01:00
parent 7418e28988
commit 9c40024a90
Signed by: Laurent
SSH key fingerprint: SHA256:kZEpW8cMJ54PDeCvOhzreNr4FSh6R13CMGH/POoO8DI
8 changed files with 181 additions and 168 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

59
src/blender/proj.py
View file

@ -1,25 +1,27 @@
import bpy import bpy
from mathutils import Matrix from mathutils import Matrix
#--------------------------------------------------------------- # ---------------------------------------------------------------
# 3x4 P matrix from Blender camera # 3x4 P matrix from Blender camera
#--------------------------------------------------------------- # ---------------------------------------------------------------
# BKE_camera_sensor_size # BKE_camera_sensor_size
def get_sensor_size(sensor_fit, sensor_x, sensor_y): def get_sensor_size(sensor_fit, sensor_x, sensor_y):
if sensor_fit == 'VERTICAL': if sensor_fit == "VERTICAL":
return sensor_y return sensor_y
return sensor_x return sensor_x
# BKE_camera_sensor_fit # BKE_camera_sensor_fit
def get_sensor_fit(sensor_fit, size_x, size_y): def get_sensor_fit(sensor_fit, size_x, size_y):
if sensor_fit == 'AUTO': if sensor_fit == "AUTO":
if size_x >= size_y: if size_x >= size_y:
return 'HORIZONTAL' return "HORIZONTAL"
else: else:
return 'VERTICAL' return "VERTICAL"
return sensor_fit return sensor_fit
# Build intrinsic camera parameters from Blender camera data # Build intrinsic camera parameters from Blender camera data
# #
# See notes on this in # See notes on this in
@ -27,8 +29,8 @@ def get_sensor_fit(sensor_fit, size_x, size_y):
# as well as # as well as
# https://blender.stackexchange.com/a/120063/3581 # https://blender.stackexchange.com/a/120063/3581
def get_calibration_matrix_K_from_blender(camd): def get_calibration_matrix_K_from_blender(camd):
if camd.type != 'PERSP': if camd.type != "PERSP":
raise ValueError('Non-perspective cameras not supported') raise ValueError("Non-perspective cameras not supported")
scene = bpy.context.scene scene = bpy.context.scene
f_in_mm = camd.lens f_in_mm = camd.lens
scale = scene.render.resolution_percentage / 100 scale = scene.render.resolution_percentage / 100
@ -38,10 +40,10 @@ def get_calibration_matrix_K_from_blender(camd):
sensor_fit = get_sensor_fit( sensor_fit = get_sensor_fit(
camd.sensor_fit, camd.sensor_fit,
scene.render.pixel_aspect_x * resolution_x_in_px, scene.render.pixel_aspect_x * resolution_x_in_px,
scene.render.pixel_aspect_y * resolution_y_in_px scene.render.pixel_aspect_y * resolution_y_in_px,
) )
pixel_aspect_ratio = scene.render.pixel_aspect_y / scene.render.pixel_aspect_x pixel_aspect_ratio = scene.render.pixel_aspect_y / scene.render.pixel_aspect_x
if sensor_fit == 'HORIZONTAL': if sensor_fit == "HORIZONTAL":
view_fac_in_px = resolution_x_in_px view_fac_in_px = resolution_x_in_px
else: else:
view_fac_in_px = pixel_aspect_ratio * resolution_y_in_px view_fac_in_px = pixel_aspect_ratio * resolution_y_in_px
@ -52,14 +54,12 @@ def get_calibration_matrix_K_from_blender(camd):
# Parameters of intrinsic calibration matrix K # Parameters of intrinsic calibration matrix K
u_0 = resolution_x_in_px / 2 - camd.shift_x * view_fac_in_px u_0 = resolution_x_in_px / 2 - camd.shift_x * view_fac_in_px
v_0 = resolution_y_in_px / 2 + camd.shift_y * view_fac_in_px / pixel_aspect_ratio v_0 = resolution_y_in_px / 2 + camd.shift_y * view_fac_in_px / pixel_aspect_ratio
skew = 0 # only use rectangular pixels skew = 0 # only use rectangular pixels
K = Matrix( K = Matrix(((s_u, skew, u_0), (0, s_v, v_0), (0, 0, 1)))
((s_u, skew, u_0),
( 0, s_v, v_0),
( 0, 0, 1)))
return K return K
# Returns camera rotation and translation matrices from Blender. # Returns camera rotation and translation matrices from Blender.
# #
# There are 3 coordinate systems involved: # There are 3 coordinate systems involved:
@ -76,10 +76,7 @@ def get_calibration_matrix_K_from_blender(camd):
# - right-handed: positive z look-at direction # - right-handed: positive z look-at direction
def get_3x4_RT_matrix_from_blender(cam): def get_3x4_RT_matrix_from_blender(cam):
# bcam stands for blender camera # bcam stands for blender camera
R_bcam2cv = Matrix( R_bcam2cv = Matrix(((1, 0, 0), (0, -1, 0), (0, 0, -1)))
((1, 0, 0),
(0, -1, 0),
(0, 0, -1)))
# Transpose since the rotation is object rotation, # Transpose since the rotation is object rotation,
# and we want coordinate rotation # and we want coordinate rotation
@ -93,29 +90,29 @@ def get_3x4_RT_matrix_from_blender(cam):
# Convert camera location to translation vector used in coordinate changes # Convert camera location to translation vector used in coordinate changes
# T_world2bcam = -1*R_world2bcam @ cam.location # T_world2bcam = -1*R_world2bcam @ cam.location
# Use location from matrix_world to account for constraints: # Use location from matrix_world to account for constraints:
T_world2bcam = -1*R_world2bcam @ location T_world2bcam = -1 * R_world2bcam @ location
# Build the coordinate transform matrix from world to computer vision camera # Build the coordinate transform matrix from world to computer vision camera
R_world2cv = R_bcam2cv@R_world2bcam R_world2cv = R_bcam2cv @ R_world2bcam
T_world2cv = R_bcam2cv@T_world2bcam T_world2cv = R_bcam2cv @ T_world2bcam
# put into 3x4 matrix # put into 3x4 matrix
RT = Matrix(( RT = Matrix(
R_world2cv[0][:] + (T_world2cv[0],), (R_world2cv[0][:] + (T_world2cv[0],), R_world2cv[1][:] + (T_world2cv[1],), R_world2cv[2][:] + (T_world2cv[2],))
R_world2cv[1][:] + (T_world2cv[1],), )
R_world2cv[2][:] + (T_world2cv[2],)
))
return RT return RT
def get_3x4_P_matrix_from_blender(cam): def get_3x4_P_matrix_from_blender(cam):
K = get_calibration_matrix_K_from_blender(cam.data) K = get_calibration_matrix_K_from_blender(cam.data)
RT = get_3x4_RT_matrix_from_blender(cam) RT = get_3x4_RT_matrix_from_blender(cam)
return K@RT, K, RT return K @ RT, K, RT
# ---------------------------------------------------------- # ----------------------------------------------------------
if __name__ == "__main__": if __name__ == "__main__":
# Insert your camera name here # Insert your camera name here
cam = bpy.data.objects['Camera'] cam = bpy.data.objects["Camera"]
P, K, RT = get_3x4_P_matrix_from_blender(cam) P, K, RT = get_3x4_P_matrix_from_blender(cam)
print("K") print("K")
print(K) print(K)
@ -132,5 +129,5 @@ if __name__ == "__main__":
# Bonus code: save the 3x4 P matrix into a plain text file # Bonus code: save the 3x4 P matrix into a plain text file
# Don't forget to import numpy for this # Don't forget to import numpy for this
#nP = numpy.matrix(P) # nP = numpy.matrix(P)
#numpy.savetxt("/tmp/P3x4.txt", nP) # to select precision, use e.g. fmt='%.2f' # numpy.savetxt("/tmp/P3x4.txt", nP) # to select precision, use e.g. fmt='%.2f'

14
src/blender/render.py
View file

@ -54,14 +54,16 @@ for i, (phi, theta) in enumerate(poses):
# save camera matrices # save camera matrices
with open(EXPORT_PATH / "cameras" / f"{i:04d}.pickle", "wb") as f: with open(EXPORT_PATH / "cameras" / f"{i:04d}.pickle", "wb") as f:
pickle.dump({ pickle.dump(
"P": np.array(P), {
"K": np.array(K), "P": np.array(P),
"RT": np.array(RT), "K": np.array(K),
}, f) "RT": np.array(RT),
},
f,
)
print(f"Saved camera matrices: {i:04d}.pickle") print(f"Saved camera matrices: {i:04d}.pickle")
# render the frame # render the frame
bpy.context.scene.frame_current = i bpy.context.scene.frame_current = i
bpy.ops.render.render(write_still=False) bpy.ops.render.render(write_still=False)

28
src/borders.py
View file

@ -25,21 +25,33 @@ def update_border(voxel_values, idx=None):
z_p1 = voxel_values[:, :, :-1] z_p1 = voxel_values[:, :, :-1]
z_p1 = np.concatenate((np.zeros((z_p1.shape[0], z_p1.shape[1], 1)), z_p1), axis=2) z_p1 = np.concatenate((np.zeros((z_p1.shape[0], z_p1.shape[1], 1)), z_p1), axis=2)
return np.logical_or.reduce((voxel_values != x_m1, voxel_values != x_p1, return np.logical_or.reduce(
voxel_values != y_m1, voxel_values != y_p1, (
voxel_values != z_m1, voxel_values != z_p1)) voxel_values != x_m1,
voxel_values != x_p1,
voxel_values != y_m1,
voxel_values != y_p1,
voxel_values != z_m1,
voxel_values != z_p1,
)
)
# TODO: update only concidered voxels (idx) # TODO: update only concidered voxels (idx)
if __name__ == '__main__': if __name__ == "__main__":
voxel_values = np.array([[[np.sqrt(x**2 + y**2 + z**2) < 20 for z in np.arange(-10, 10, 1.0)] for y in np.arange(-10, 10, 1.0)] for x in np.arange(-10, 10, 1.0)]) voxel_values = np.array(
[
[[np.sqrt(x**2 + y**2 + z**2) < 20 for z in np.arange(-10, 10, 1.0)] for y in np.arange(-10, 10, 1.0)]
for x in np.arange(-10, 10, 1.0)
]
)
border = update_border(voxel_values) border = update_border(voxel_values)
# Plot voxel grid that are the border # Plot voxel grid that are the border
fig = plt.figure() fig = plt.figure()
ax = fig.add_subplot(111, projection='3d') ax = fig.add_subplot(111, projection="3d")
ax.scatter(np.where(voxel_values)[0], np.where(voxel_values)[1], np.where(voxel_values)[2], c='r', marker='o') ax.scatter(np.where(voxel_values)[0], np.where(voxel_values)[1], np.where(voxel_values)[2], c="r", marker="o")
ax.scatter(np.where(border)[0], np.where(border)[1], np.where(border)[2], c='b', marker='o', s=1) ax.scatter(np.where(border)[0], np.where(border)[1], np.where(border)[2], c="b", marker="o", s=1)
plt.show() plt.show()

60
src/fvi.py
View file

@ -53,8 +53,9 @@ def fast_voxel_intersect(start, end, origin, step, shape) -> tuple[list, list, l
next_boundaries = np.divide(position + step * direction_signs, step) next_boundaries = np.divide(position + step * direction_signs, step)
errored = np.abs(np.round(next_boundaries) - next_boundaries) < 1e-12 errored = np.abs(np.round(next_boundaries) - next_boundaries) < 1e-12
next_boundaries[errored] = np.round(next_boundaries[errored]) next_boundaries[errored] = np.round(next_boundaries[errored])
distances = ((1 - is_negative) * np.floor(next_boundaries) + distances = (
is_negative * np.ceil(next_boundaries)) * step - position (1 - is_negative) * np.floor(next_boundaries) + is_negative * np.ceil(next_boundaries)
) * step - position
# Determine the nearest boundary to be reached # Determine the nearest boundary to be reached
boundary_distances = np.abs(distances / direction) boundary_distances = np.abs(distances / direction)
@ -71,8 +72,9 @@ def fast_voxel_intersect(start, end, origin, step, shape) -> tuple[list, list, l
# print("position_update: ", position) # print("position_update: ", position)
# Correct position to be on boundary # Correct position to be on boundary
position[clothest_boundary] = round( position[clothest_boundary] = (
position[clothest_boundary] / step[clothest_boundary]) * step[clothest_boundary] round(position[clothest_boundary] / step[clothest_boundary]) * step[clothest_boundary]
)
# Get corresponding voxel # Get corresponding voxel
on_boundary = np.mod(position, step) == 0 on_boundary = np.mod(position, step) == 0
@ -94,7 +96,7 @@ def fast_voxel_intersect(start, end, origin, step, shape) -> tuple[list, list, l
return intersections, voxels, voxels_idx return intersections, voxels, voxels_idx
if __name__ == '__main__': if __name__ == "__main__":
import matplotlib.pyplot as plt import matplotlib.pyplot as plt
def update_figure(): def update_figure():
@ -104,30 +106,42 @@ if __name__ == '__main__':
# Plot hitted voxels # Plot hitted voxels
for voxel in voxels: for voxel in voxels:
plt.fill([voxel[0], voxel[0] + step[0], voxel[0] + step[0], voxel[0]], plt.fill(
[voxel[1], voxel[1], voxel[1] + step[1], voxel[1] + step[1]], [voxel[0], voxel[0] + step[0], voxel[0] + step[0], voxel[0]],
color='#e25', alpha=0.5) [voxel[1], voxel[1], voxel[1] + step[1], voxel[1] + step[1]],
color="#e25",
alpha=0.5,
)
for voxel_id in voxels_idx: for voxel_id in voxels_idx:
plt.fill([ plt.fill(
origin[0] + voxel_id[0] * step[0], origin[0] + (voxel_id[0] + 1) * step[0], [
origin[0] + (voxel_id[0] + 1) * step[0], origin[0] + voxel_id[0] * step[0] origin[0] + voxel_id[0] * step[0],
], [ origin[0] + (voxel_id[0] + 1) * step[0],
origin[1] + voxel_id[1] * step[1], origin[1] + voxel_id[1] * step[1], origin[0] + (voxel_id[0] + 1) * step[0],
origin[1] + (voxel_id[1] + 1) * step[1], origin[1] + (voxel_id[1] + 1) * step[1] origin[0] + voxel_id[0] * step[0],
], color='#2e3', alpha=0.5) ],
[
origin[1] + voxel_id[1] * step[1],
origin[1] + voxel_id[1] * step[1],
origin[1] + (voxel_id[1] + 1) * step[1],
origin[1] + (voxel_id[1] + 1) * step[1],
],
color="#2e3",
alpha=0.5,
)
# Plot line segment # Plot line segment
plt.plot([start[0], end[0]], [start[1], end[1]], 'k-') plt.plot([start[0], end[0]], [start[1], end[1]], "k-")
plt.plot(start[0], start[1], 'go') plt.plot(start[0], start[1], "go")
plt.plot(end[0], end[1], 'ro') plt.plot(end[0], end[1], "ro")
# Plot intersection points # Plot intersection points
for pos in positions: for pos in positions:
plt.plot(pos[0], pos[1], 'bo') plt.plot(pos[0], pos[1], "bo")
# Plot voxel grid # Plot voxel grid
plt.axis('equal') plt.axis("equal")
plt.xlim((-10, 10)) plt.xlim((-10, 10))
plt.ylim((-10, 10)) plt.ylim((-10, 10))
plt.xticks(origin[0] + step[0] * np.arange(shape[0] + 1)) plt.xticks(origin[0] + step[0] * np.arange(shape[0] + 1))
@ -137,13 +151,13 @@ if __name__ == '__main__':
def onkey(event): def onkey(event):
global start, end global start, end
if event.key == ' ': if event.key == " ":
start = np.random.rand(2) * 20 - 10 start = np.random.rand(2) * 20 - 10
end = np.random.rand(2) * 20 - 10 end = np.random.rand(2) * 20 - 10
update_figure() update_figure()
# Define voxel grid # Define voxel grid
origin = np.array([-5., -5.]) origin = np.array([-5.0, -5.0])
step = np.array([0.7, 0.7]) step = np.array([0.7, 0.7])
shape = (10, 10) shape = (10, 10)
@ -155,6 +169,6 @@ if __name__ == '__main__':
# Plot # Plot
fig = plt.figure() fig = plt.figure()
fig.canvas.mpl_connect('key_press_event', onkey) fig.canvas.mpl_connect("key_press_event", onkey)
update_figure() update_figure()
plt.show() plt.show()

92
src/intersec_line_voxel.py
View file

@ -1,14 +1,10 @@
import numpy as np
import matplotlib.pyplot as plt
from itertools import product from itertools import product
import matplotlib.pyplot as plt
import numpy as np
def check_line_voxel(
px, py, pz, def check_line_voxel(px, py, pz, dx, dy, dz, vx, vy, vz, c):
dx, dy, dz,
vx, vy, vz,
c
):
"""Check if a line intersects a voxel. """Check if a line intersects a voxel.
Parameters: Parameters:
@ -17,7 +13,6 @@ def check_line_voxel(
- vx, vy, vz: line direction coordinates - vx, vy, vz: line direction coordinates
- c: voxel size - c: voxel size
""" """
# Compute the intersection bounds # Compute the intersection bounds
kx1 = (px - dx) / vx kx1 = (px - dx) / vx
ky1 = (py - dy) / vy ky1 = (py - dy) / vy
@ -27,54 +22,27 @@ def check_line_voxel(
kz2 = (pz - dz + c) / vz kz2 = (pz - dz + c) / vz
# Order the bounds # Order the bounds
kxmin = np.min(np.concatenate([ kxmin = np.min(np.concatenate([kx1[:, np.newaxis], kx2[:, np.newaxis]], axis=1), axis=1)
kx1[:, np.newaxis], kymin = np.min(np.concatenate([ky1[:, np.newaxis], ky2[:, np.newaxis]], axis=1), axis=1)
kx2[:, np.newaxis] kzmin = np.min(np.concatenate([kz1[:, np.newaxis], kz2[:, np.newaxis]], axis=1), axis=1)
], axis=1), axis=1) kxmax = np.max(np.concatenate([kx1[:, np.newaxis], kx2[:, np.newaxis]], axis=1), axis=1)
kymin = np.min(np.concatenate([ kymax = np.max(np.concatenate([ky1[:, np.newaxis], ky2[:, np.newaxis]], axis=1), axis=1)
ky1[:, np.newaxis], kzmax = np.max(np.concatenate([kz1[:, np.newaxis], kz2[:, np.newaxis]], axis=1), axis=1)
ky2[:, np.newaxis]
], axis=1), axis=1)
kzmin = np.min(np.concatenate([
kz1[:, np.newaxis],
kz2[:, np.newaxis]
], axis=1), axis=1)
kxmax = np.max(np.concatenate([
kx1[:, np.newaxis],
kx2[:, np.newaxis]
], axis=1), axis=1)
kymax = np.max(np.concatenate([
ky1[:, np.newaxis],
ky2[:, np.newaxis]
], axis=1), axis=1)
kzmax = np.max(np.concatenate([
kz1[:, np.newaxis],
kz2[:, np.newaxis]
], axis=1), axis=1)
# Check if the bounds overlap # Check if the bounds overlap
kmax = np.min(np.concatenate([ kmax = np.min(np.concatenate([kxmax[:, np.newaxis], kymax[:, np.newaxis], kzmax[:, np.newaxis]], axis=1), axis=1)
kxmax[:, np.newaxis], kmin = np.max(np.concatenate([kxmin[:, np.newaxis], kymin[:, np.newaxis], kzmin[:, np.newaxis]], axis=1), axis=1)
kymax[:, np.newaxis],
kzmax[:, np.newaxis]
], axis=1), axis=1)
kmin = np.max(np.concatenate([
kxmin[:, np.newaxis],
kymin[:, np.newaxis],
kzmin[:, np.newaxis]
], axis=1), axis=1)
return kmin <= kmax return kmin <= kmax
c = 1.0 c = 1.0
points = np.array([[x, y, z] for x, y, z in product( points = np.array(
np.arange(-5.0, 4.0, c), [[x, y, z] for x, y, z in product(np.arange(-5.0, 4.0, c), np.arange(-5.0, 4.0, c), np.arange(-5.0, 4.0, c))]
np.arange(-5.0, 4.0, c), )
np.arange(-5.0, 4.0, c))
])
while True: while True:
fig = plt.figure() fig = plt.figure()
ax = plt.axes(projection='3d') ax = plt.axes(projection="3d")
d = np.random.rand(3) * 1 - 0.5 d = np.random.rand(3) * 1 - 0.5
v = np.random.rand(3) * 1 - 0.5 v = np.random.rand(3) * 1 - 0.5
@ -90,20 +58,20 @@ while True:
if not bool_vect[i]: if not bool_vect[i]:
continue continue
ax.plot([px, px+c], [py, py], [pz, pz], 'b') ax.plot([px, px + c], [py, py], [pz, pz], "b")
ax.plot([px, px+c], [py, py], [pz+c, pz+c], 'b') ax.plot([px, px + c], [py, py], [pz + c, pz + c], "b")
ax.plot([px, px+c], [py+c, py+c], [pz, pz], 'b') ax.plot([px, px + c], [py + c, py + c], [pz, pz], "b")
ax.plot([px, px+c], [py+c, py+c], [pz+c, pz+c], 'b') ax.plot([px, px + c], [py + c, py + c], [pz + c, pz + c], "b")
ax.plot([px, px], [py, py+c], [pz, pz], 'b') ax.plot([px, px], [py, py + c], [pz, pz], "b")
ax.plot([px, px], [py, py+c], [pz+c, pz+c], 'b') ax.plot([px, px], [py, py + c], [pz + c, pz + c], "b")
ax.plot([px+c, px+c], [py, py+c], [pz, pz], 'b') ax.plot([px + c, px + c], [py, py + c], [pz, pz], "b")
ax.plot([px+c, px+c], [py, py+c], [pz+c, pz+c], 'b') ax.plot([px + c, px + c], [py, py + c], [pz + c, pz + c], "b")
ax.plot([px, px], [py, py], [pz, pz+c], 'b') ax.plot([px, px], [py, py], [pz, pz + c], "b")
ax.plot([px, px], [py+c, py+c], [pz, pz+c], 'b') ax.plot([px, px], [py + c, py + c], [pz, pz + c], "b")
ax.plot([px+c, px+c], [py, py], [pz, pz+c], 'b') ax.plot([px + c, px + c], [py, py], [pz, pz + c], "b")
ax.plot([px+c, px+c], [py+c, py+c], [pz, pz+c], 'b') ax.plot([px + c, px + c], [py + c, py + c], [pz, pz + c], "b")
# plot line # plot line
ax.plot([dx, dx+vx], [dy, dy+vy], [dz, dz+vz], 'g') ax.plot([dx, dx + vx], [dy, dy + vy], [dz, dz + vz], "g")
plt.show() plt.show()

63
src/main.py
View file

@ -17,10 +17,16 @@ Z_MIN, Z_MAX = -0.1, 0.1
nb_frame = 24 nb_frame = 24
points = np.array([[x, y, z, 1.0] for x, y, z in product( points = np.array(
np.arange(X_MIN, X_MAX, VOXEL_SIZE), [
np.arange(Y_MIN, Y_MAX, VOXEL_SIZE), [x, y, z, 1.0]
np.arange(Z_MIN, Z_MAX, VOXEL_SIZE))]) for x, y, z in product(
np.arange(X_MIN, X_MAX, VOXEL_SIZE),
np.arange(Y_MIN, Y_MAX, VOXEL_SIZE),
np.arange(Z_MIN, Z_MAX, VOXEL_SIZE),
)
]
)
mask = 255 mask = 255
@ -29,10 +35,10 @@ proj_mats = []
frames = [] frames = []
for k in range(nb_frame): for k in range(nb_frame):
frame = cv2.imread(f'data/torus/masks/Image{k:04}.png', cv2.IMREAD_GRAYSCALE) frame = cv2.imread(f"data/torus/masks/Image{k:04}.png", cv2.IMREAD_GRAYSCALE)
frames.append(cv2.imread(f'data/torus/images/Image{k:04}.png', cv2.IMREAD_GRAYSCALE)) frames.append(cv2.imread(f"data/torus/images/Image{k:04}.png", cv2.IMREAD_GRAYSCALE))
with open(f"data/torus/cameras/{k:04d}.pickle", 'rb') as file: with open(f"data/torus/cameras/{k:04d}.pickle", "rb") as file:
matrices = pickle.load(file) matrices = pickle.load(file)
proj_mat = matrices["P"] proj_mat = matrices["P"]
proj_mats.append(proj_mat) proj_mats.append(proj_mat)
@ -40,16 +46,23 @@ for k in range(nb_frame):
positions.append(position) positions.append(position)
cam_points = proj_mat @ points.T cam_points = proj_mat @ points.T
cam_points /= cam_points[2,:] cam_points /= cam_points[2, :]
cam_points = np.round(cam_points).astype(np.int32) cam_points = np.round(cam_points).astype(np.int32)
visible = np.logical_and.reduce((0 <= cam_points[0,:], cam_points[0,:] < frame.shape[1], 0 <= cam_points[1,:], cam_points[1,:] < frame.shape[0])) visible = np.logical_and.reduce(
cam_points = cam_points[:,visible] (
points = points[visible,:] 0 <= cam_points[0, :],
cam_points[0, :] < frame.shape[1],
0 <= cam_points[1, :],
cam_points[1, :] < frame.shape[0],
)
)
cam_points = cam_points[:, visible]
points = points[visible, :]
solid = frame[cam_points[1,:],cam_points[0,:]] == mask solid = frame[cam_points[1, :], cam_points[0, :]] == mask
cam_points = cam_points[:,solid] cam_points = cam_points[:, solid]
points = points[solid,:] points = points[solid, :]
# for cam_point in cam_points.T: # for cam_point in cam_points.T:
# cv2.circle(frame, (cam_point[0], cam_point[1]), 2, (255*is_in[k], 0, 255*(not is_in[k]))) # cv2.circle(frame, (cam_point[0], cam_point[1]), 2, (255*is_in[k], 0, 255*(not is_in[k])))
@ -73,9 +86,15 @@ for k in range(nb_frame):
# cv2.waitKey(0) # cv2.waitKey(0)
voxel = np.zeros((int((X_MAX-X_MIN)/VOXEL_SIZE + 1), int((Y_MAX-Y_MIN)/VOXEL_SIZE + 1), int((Z_MAX-Z_MIN)/VOXEL_SIZE + 1))) voxel = np.zeros(
(
int((X_MAX - X_MIN) / VOXEL_SIZE + 1),
int((Y_MAX - Y_MIN) / VOXEL_SIZE + 1),
int((Z_MAX - Z_MIN) / VOXEL_SIZE + 1),
)
)
idx = np.floor_divide(points[:, :3] - np.array([X_MIN, Y_MIN, Z_MIN]), VOXEL_SIZE).astype(int) idx = np.floor_divide(points[:, :3] - np.array([X_MIN, Y_MIN, Z_MIN]), VOXEL_SIZE).astype(int)
voxel[idx[:,0], idx[:,1], idx[:,2]] = 1 voxel[idx[:, 0], idx[:, 1], idx[:, 2]] = 1
border = update_border(voxel) border = update_border(voxel)
@ -88,14 +107,15 @@ border = update_border(voxel)
origin = np.array([X_MIN, Y_MIN, Z_MIN]) origin = np.array([X_MIN, Y_MIN, Z_MIN])
step = np.array([VOXEL_SIZE, VOXEL_SIZE, VOXEL_SIZE]) step = np.array([VOXEL_SIZE, VOXEL_SIZE, VOXEL_SIZE])
shape = np.array([int((X_MAX-X_MIN)/VOXEL_SIZE), int((Y_MAX-Y_MIN)/VOXEL_SIZE), int((Z_MAX-Z_MIN)/VOXEL_SIZE)]) shape = np.array(
[int((X_MAX - X_MIN) / VOXEL_SIZE), int((Y_MAX - Y_MIN) / VOXEL_SIZE), int((Z_MAX - Z_MIN) / VOXEL_SIZE)]
)
for idx in track(np.argwhere(border)): for idx in track(np.argwhere(border)):
# coordonnées du centre du voxel # coordonnées du centre du voxel
start = np.array([ start = np.array(
X_MIN + (idx[0] + 0.5) * VOXEL_SIZE, [X_MIN + (idx[0] + 0.5) * VOXEL_SIZE, Y_MIN + (idx[1] + 0.5) * VOXEL_SIZE, Z_MIN + (idx[2] + 0.5) * VOXEL_SIZE]
Y_MIN + (idx[1] + 0.5) * VOXEL_SIZE, )
Z_MIN + (idx[2] + 0.5) * VOXEL_SIZE])
# array qui contiendra les nuances de gris des frames qui voient le voxel # array qui contiendra les nuances de gris des frames qui voient le voxel
values = [] values = []
@ -130,6 +150,5 @@ for idx in track(np.argwhere(border)):
voxel[idx[0], idx[1], idx[2]] = 0 voxel[idx[0], idx[1], idx[2]] = 0
vertices, triangles = mcubes.marching_cubes(voxel, 0) vertices, triangles = mcubes.marching_cubes(voxel, 0)
mcubes.export_obj(vertices, triangles, "result.obj") mcubes.export_obj(vertices, triangles, "result.obj")

14
src/matrices_reader.py
View file

@ -12,23 +12,25 @@ def matrices_reader(path: str) -> list[np.ndarray]:
list[np.ndarray]: list of projection matrix list[np.ndarray]: list of projection matrix
""" """
with open(path, 'r') as f: with open(path, "r") as f:
lines = f.readlines() lines = f.readlines()
k = 0 k = 0
world_matrices = [] world_matrices = []
while k+3 < len(lines): while k + 3 < len(lines):
# Match matrices one by one # Match matrices one by one
mat_str = "" mat_str = ""
for line in lines[k:k+4]: for line in lines[k : k + 4]:
mat_str += line mat_str += line
float_reg = r"(-|\d|\.|e)+" float_reg = r"(-|\d|\.|e)+"
res = re.search( res = re.search(
f"Matrix\(\(\(({float_reg}), ({float_reg}), ({float_reg}), ({float_reg})\),\n +\(({float_reg}), ({float_reg}), ({float_reg}), ({float_reg})\),\n\ +\(({float_reg}), ({float_reg}), ({float_reg}), ({float_reg})\)\)\)", mat_str) f"Matrix\(\(\(({float_reg}), ({float_reg}), ({float_reg}), ({float_reg})\),\n +\(({float_reg}), ({float_reg}), ({float_reg}), ({float_reg})\),\n\ +\(({float_reg}), ({float_reg}), ({float_reg}), ({float_reg})\)\)\)",
mat_str,
)
# Convert string to np.ndarray # Convert string to np.ndarray
values = [float(res.group(i)) for i in range(1,len(res.groups()) + 1, 2)] values = [float(res.group(i)) for i in range(1, len(res.groups()) + 1, 2)]
world_mat = np.array([[values[4*i + j] for j in range(4)] for i in range(3)]) world_mat = np.array([[values[4 * i + j] for j in range(4)] for i in range(3)])
world_matrices.append(world_mat) world_matrices.append(world_mat)
k += 4 k += 4

3
src/test.py
View file

@ -4,8 +4,7 @@ nb_frame = 24
for k in range(nb_frame): for k in range(nb_frame):
with open(f"/tmp/cameras/{k:04d}.pickle", 'rb') as file: with open(f"/tmp/cameras/{k:04d}.pickle", "rb") as file:
proj_mat = pickle.load(file)["P"] proj_mat = pickle.load(file)["P"]
print(k, proj_mat) print(k, proj_mat)