Skip to content

P
py-3d-face-reconstruction
Commit c2204dea authored by Richard Torenvliet's avatar Richard Torenvliet
Browse files
Options

Moved from the use of python solve, to simply move to use multiplication 

It's now around 3x faster then before
parent cb6d90c9
Hide whitespace changes
Inline Side-by-side
Showing with 188 additions and 11311 deletions
makefile
View file @ c2204dea
......@@ -12,7 +12,7 @@ include actions.mk
include src/reconstruction/build.mk
data: data/imm_face_db
reconstruction: texture.so halide src/reconstruction/texture_halide
reconstruction: texture.so src/reconstruction/texture_halide
OS := $(shell uname)
......
src/reconstruction/texture.c deleted 100644 → 0
View file @ cb6d90c9
This source diff could not be displayed because it is too large. You can view the blob instead.
src/reconstruction/texture.pyx
View file @ c2204dea
......@@ -2,6 +2,7 @@ cimport cython
import numpy as np
cimport numpy as np
from cpython cimport array as c_array
from cython.view cimport array as cvarray
DTYPE_int = np.int
ctypedef np.int_t DTYPE_int_t
......@@ -11,6 +12,7 @@ DTYPE_float64 = np.float64
ctypedef np.float32_t DTYPE_float32_t
ctypedef np.float64_t DTYPE_float64_t
ctypedef float * float_ptr
cdef inline float cross_product(int v1_x, int v1_y, int v2_x, int v2_y):
......@@ -18,8 +20,7 @@ cdef inline float cross_product(int v1_x, int v1_y, int v2_x, int v2_y):
return (v1_x * v2_y) - (v1_y * v2_x)
cdef inline np.ndarray[double, ndim=1] cartesian2barycentric(
int r1_x, r1_y, int r2_x, int r2_y, int r3_x, int r3_y, int r_x, int r_y):
def cartesian2barycentric_slow_test(int r1_x, r1_y, int r2_x, int r2_y, int r3_x, int r3_y, int r_x, int r_y):
"""
Given a triangle spanned by three cartesion points
r1, r2, r2, and point r, return the barycentric weights l1, l2, l3.
......@@ -39,6 +40,84 @@ cdef inline np.ndarray[double, ndim=1] cartesian2barycentric(
return np.linalg.solve(a, b)
def cartesian2barycentric_test(
float x_1, float y_1, float x_2, float y_2, float x_3, float y_3, float x, float y):
"""
lambda_1 = (y_2 - y_3)(x - x_3) + (x_3 - x_2)(y - y_3) /
(y_2-y_3)(x_1-x_3)+(x_3-x_2)(y_1-y_3)
lambda_2 = (y_3 - y_1)(x - x_3) + (x_1 - x_3)(y - y_3) /
(y_2-y_3)(x_1-x_3)+(x_3-x_2)(y_1-y_3)
lambda_3 = 1 lambda_1 - lambda_2
Returns:
ndarray (of dim 3) weights of the barycentric coordinates
"""
cdef float lambda_1 = ((y_2 - y_3) * (x - x_3) + (x_3 - x_2) * (y - y_3)) / \
((y_2 - y_3) * (x_1 - x_3) + (x_3 - x_2) * (y_1 - y_3))
cdef float lambda_2 = ((y_3 - y_1) * (x - x_3) + (x_1 - x_3) * (y - y_3)) / \
((y_2 - y_3) * (x_1 - x_3) + (x_3 - x_2) * (y_1 - y_3))
cdef float lambda_3 = 1 - lambda_1 - lambda_2
return [lambda_1, lambda_2, lambda_3]
cdef inline cartesian2barycentric(
float x_1, float y_1, float x_2, float y_2, float x_3, float y_3, float
x, float y, float[:] lambdas):
"""
lambda_1 = (y_2 - y_3)(x - x_3) + (x_3 - x_2)(y - y_3) /
(y_2-y_3)(x_1-x_3)+(x_3-x_2)(y_1-y_3)
lambda_2 = (y_3 - y_1)(x - x_3) + (x_1 - x_3)(y - y_3) /
(y_2-y_3)(x_1-x_3)+(x_3-x_2)(y_1-y_3)
lambda_3 = 1 lambda_1 - lambda_2
Returns:
ndarray (of dim 3) weights of the barycentric coordinates
"""
lambdas[0] = ((y_2 - y_3) * (x - x_3) + (x_3 - x_2) * (y - y_3)) / \
((y_2 - y_3) * (x_1 - x_3) + (x_3 - x_2) * (y_1 - y_3))
lambdas[1] = ((y_3 - y_1) * (x - x_3) + (x_1 - x_3) * (y - y_3)) / \
((y_2 - y_3) * (x_1 - x_3) + (x_3 - x_2) * (y_1 - y_3))
#cdef float lambda_3 = 1 - lambda_1 - lambda_2
#a = np.array([
# [r1_x, r2_x, r3_x],
# [r1_y, r2_y, r3_y],
# [1, 1, 1]
#])
#b = np.array([r_x, r_y, 1])
#return np.linalg.solve(a, b)
cdef inline np.ndarray[double, ndim=1] cartesian2barycentric_slow(
int r1_x, r1_y, int r2_x, int r2_y, int r3_x, int r3_y, int
r_x, int r_y):
"""
Given a triangle spanned by three cartesion points
r1, r2, r2, and point r, return the barycentric weights l1, l2, l3.
Returns:
ndarray (of dim 3) weights of the barycentric coordinates
"""
a = np.array([
[r1_x, r2_x, r3_x],
[r1_y, r2_y, r3_y],
[1, 1, 1]
])
b = np.array([r_x, r_y, 1])
return np.linalg.solve(a, b)
cdef inline np.ndarray[double, ndim=2] barycentric2cartesian(
int x1, int x2, int x3, int y1, int y2, int y3,
......@@ -75,18 +154,17 @@ def fill_triangle(np.ndarray[unsigned char, ndim=3] src,
cdef np.ndarray matrix = np.full([3, 3], fill_value=1, dtype=DTYPE_int)
cdef np.ndarray src_loc = np.zeros([3, 1], dtype=DTYPE_float64)
cdef np.ndarray dst_loc = np.zeros([3, 1], dtype=DTYPE_float64)
cdef int w = x_max - x_min
cdef int h = y_max - y_min
cdef int new_offset
#cdef np.ndarray src_reshaped = src[offset:offset + (w * h * 3)].reshape((h, w, 3))
cdef c_array.array dst_loc = c_array.array('f', [0., 0.])
for j, y in enumerate(xrange(y_min, y_max)):
for i, x in enumerate(xrange(x_min, x_max)):
dst_loc = cartesian2barycentric(
x1, y1, x2, y2, x3, y3, x, y
cartesian2barycentric(
x1, y1, x2, y2, x3, y3, x, y, dst_loc
)
s = dst_loc[0]
......@@ -98,53 +176,53 @@ def fill_triangle(np.ndarray[unsigned char, ndim=3] src,
dst[y, x, :] = src[y, x, :]
@cython.boundscheck(False)
@cython.wraparound(False)
def fill_triangle_src_dst(np.ndarray[unsigned char, ndim=3] src,
np.ndarray[unsigned char, ndim=3] dst,
int src_x1, int src_y1, int src_x2, int src_y2,
int src_x3, int src_y3,
int dst_x1, int dst_y1, int dst_x2, int dst_y2,
int dst_x3, int dst_y3,
int offset_x, int offset_y):
cdef np.ndarray triangle_x = np.array([dst_x1, dst_x2, dst_x3])
cdef np.ndarray triangle_y = np.array([dst_y1, dst_y2, dst_y3])
cdef int x_min = np.argmin(triangle_x)
cdef int x_max = np.argmax(triangle_x)
cdef int y_min = np.argmin(triangle_y)
cdef int y_max = np.argmax(triangle_y)
cdef np.ndarray L = np.zeros([3, 1], dtype=DTYPE_float32)
cdef np.ndarray matrix = np.full([3, 3], fill_value=1, dtype=DTYPE_int)
cdef np.ndarray src_loc = np.zeros([3, 1], dtype=DTYPE_float64)
cdef np.ndarray dst_loc = np.zeros([3, 1], dtype=DTYPE_float64)
for j, y in enumerate(xrange(triangle_y[y_min], triangle_y[y_max])):
for i, x in enumerate(xrange(triangle_x[x_min], triangle_x[x_max])):
dst_loc = cartesian2barycentric(
dst_x1, dst_y1, dst_x2, dst_y2, dst_x3, dst_y3, x, y
)
s = dst_loc[0]
t = dst_loc[1]
if s >= -0.001 and t >= -0.001 and s + t <= 1.001:
L[0] = s
L[1] = t
L[2] = 1 - s - t
src_loc = barycentric2cartesian(
src_x1, src_x2, src_x3,
src_y1, src_y2, src_y3,
matrix,
L
)
dst[y, x, :] = src[src_loc[1][0], src_loc[0][0], :]
#@cython.boundscheck(False)
#@cython.wraparound(False)
#def fill_triangle_src_dst(np.ndarray[unsigned char, ndim=3] src,
# np.ndarray[unsigned char, ndim=3] dst,
# int src_x1, int src_y1, int src_x2, int src_y2,
# int src_x3, int src_y3,
# int dst_x1, int dst_y1, int dst_x2, int dst_y2,
# int dst_x3, int dst_y3,
# int offset_x, int offset_y):
#
# cdef np.ndarray triangle_x = np.array([dst_x1, dst_x2, dst_x3])
# cdef np.ndarray triangle_y = np.array([dst_y1, dst_y2, dst_y3])
#
# cdef int x_min = np.argmin(triangle_x)
# cdef int x_max = np.argmax(triangle_x)
# cdef int y_min = np.argmin(triangle_y)
# cdef int y_max = np.argmax(triangle_y)
#
# cdef np.ndarray L = np.zeros([3, 1], dtype=DTYPE_float32)
# cdef np.ndarray matrix = np.full([3, 3], fill_value=1, dtype=DTYPE_int)
#
# cdef np.ndarray src_loc = np.zeros([3, 1], dtype=DTYPE_float64)
# cdef c_array.array dst_loc = c_array.array('f', [0., 0.])
#
# for j, y in enumerate(xrange(triangle_y[y_min], triangle_y[y_max])):
# for i, x in enumerate(xrange(triangle_x[x_min], triangle_x[x_max])):
# cartesian2barycentric(
# dst_x1, dst_y1, dst_x2, dst_y2, dst_x3, dst_y3, x, y, dst_loc
# )
#
# s = dst_loc[0]
# t = dst_loc[1]
#
# if s >= -0.001 and t >= -0.001 and s + t <= 1.001:
# L[0] = s
# L[1] = t
# L[2] = 1 - s - t
#
# src_loc = barycentric2cartesian(
# src_x1, src_x2, src_x3,
# src_y1, src_y2, src_y3,
# matrix,
# L
# )
#
# dst[y, x, :] = src[src_loc[1][0], src_loc[0][0], :]
#
@cython.boundscheck(False)
@cython.wraparound(False)
......@@ -170,15 +248,15 @@ def fill_triangle_src_dst(np.ndarray[unsigned char, ndim=3] src,
cdef np.ndarray matrix = np.full([3, 3], fill_value=1, dtype=DTYPE_int)
cdef np.ndarray src_loc = np.zeros([3, 1], dtype=DTYPE_float64)
cdef np.ndarray dst_loc = np.zeros([3, 1], dtype=DTYPE_float64)
cdef c_array.array dst_loc = c_array.array('f', [0., 0.])
for j, y in enumerate(xrange(triangle_y[y_min], triangle_y[y_max])):
for i, x in enumerate(xrange(triangle_x[x_min], triangle_x[x_max])):
dst_loc = cartesian2barycentric(
cartesian2barycentric(
triangle_x[0], triangle_y[0],
triangle_x[1], triangle_y[1],
triangle_x[2], triangle_y[2],
x, y
x, y, dst_loc
)
s = dst_loc[0]
......
src/reconstruction/texture_test.py
View file @ c2204dea
import cv2
import aam
import numpy as np
from .texture import cartesian2barycentric_slow_test, cartesian2barycentric_test
def test_cartesian2barycentricy():
blue_points = [[20, 20], [50, 160], [160, 20]]
x_1 = blue_points[0][0]
y_1 = blue_points[0][1]
x_2 = blue_points[1][0]
y_2 = blue_points[1][1]
x_3 = blue_points[2][0]
y_3 = blue_points[2][1]
x = 50
y = 70
lambdas_quick = cartesian2barycentric_test(x_1, y_1, x_2, y_2, x_3, y_3, x, y)
lambdas_quick = np.asarray(lambdas_quick)
lambdas_slow = cartesian2barycentric_slow_test(x_1, y_1, x_2, y_2, x_3, y_3, x, y)
np.testing.assert_array_equal(lambdas_quick, lambdas_slow)
def test_sample_from_triangles():
......@@ -16,32 +40,32 @@ def test_sample_from_triangles():
triangles = [[0, 1, 2]]
for tri in triangles:
cv2.line(blue_image,
tuple(blue_points[tri[0]]),
tuple(blue_points[tri[1]]), (0, 255, 0), 1)
cv2.line(blue_image,
tuple(blue_points[tri[1]]),
tuple(blue_points[tri[2]]), (0, 255, 0), 1)
cv2.line(blue_image,
tuple(blue_points[tri[2]]),
tuple(blue_points[tri[0]]), (0, 255, 0), 1)
for tri in triangles:
cv2.line(red_image,
tuple(red_points[tri[0]]),
tuple(red_points[tri[1]]), (0, 255, 0), 1)
cv2.line(red_image,
tuple(red_points[tri[1]]),
tuple(red_points[tri[2]]), (0, 255, 0), 1)
cv2.line(red_image,
tuple(red_points[tri[2]]),
tuple(red_points[tri[0]]), (0, 255, 0), 1)
all_triangles = aam.sample_from_triangles(
red_image, red_points, triangles
)
cv2.imshow('blue_image', blue_image)
cv2.imshow('red_image', red_image)
cv2.waitKey(0)
#for tri in triangles:
# cv2.line(blue_image,
# tuple(blue_points[tri[0]]),
# tuple(blue_points[tri[1]]), (0, 255, 0), 1)
# cv2.line(blue_image,
# tuple(blue_points[tri[1]]),
# tuple(blue_points[tri[2]]), (0, 255, 0), 1)
# cv2.line(blue_image,
# tuple(blue_points[tri[2]]),
# tuple(blue_points[tri[0]]), (0, 255, 0), 1)
#for tri in triangles:
# cv2.line(red_image,
# tuple(red_points[tri[0]]),
# tuple(red_points[tri[1]]), (0, 255, 0), 1)
# cv2.line(red_image,
# tuple(red_points[tri[1]]),
# tuple(red_points[tri[2]]), (0, 255, 0), 1)
# cv2.line(red_image,
# tuple(red_points[tri[2]]),
# tuple(red_points[tri[0]]), (0, 255, 0), 1)
#all_triangles = aam.sample_from_triangles(
# red_image, red_points, triangles
#)
#cv2.imshow('blue_image', blue_image)
#cv2.imshow('red_image', red_image)
#cv2.waitKey(0)
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment