atomizer/src/atom/frame3.py at main · xavierchermain/atomizer · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
from math import prod

import numpy as np
import taichi as ti

from . import cone, direction, grid3


class Field:
    def __init__(self):
        self.grid = None
        self.point = None
        self.normal = None
        self.phi_t = None

    def compute_memory_usage(self):
        # Return the number of bytes required to store the frame field

        # 6 * 32 bits:
        # - Point: 3 * 32 bits
        # - Spherical normal: 2 * 32 bits
        # - Tangent angle: 1 * 32 bits

        # Unit: bytes (that's why there is * 4)
        return prod(self.grid.cell_3dcount) * 6.0 * 4.0

    def to_numpy(self):
        grid_np = self.grid.to_numpy()
        point_np = self.point.to_numpy()
        normal_np = self.normal.to_numpy()
        phi_t_np = self.phi_t.to_numpy()

        dict_array = {}
        dict_array["grid"] = grid_np
        dict_array["point"] = point_np
        dict_array["normal"] = normal_np
        dict_array["phi_t"] = phi_t_np

        return dict_array

    def from_numpy(self, dict_array):
        self.grid = grid3.Grid()
        self.grid.from_numpy(dict_array["grid"])

        shape = dict_array["phi_t"].shape

        self.point = ti.Vector.field(n=3, dtype=ti.f32, shape=shape)
        self.normal = ti.Vector.field(n=2, dtype=ti.f32, shape=shape)
        self.phi_t = ti.field(dtype=ti.f32, shape=shape)

        self.point.from_numpy(dict_array["point"])
        self.normal.from_numpy(dict_array["normal"])
        self.phi_t.from_numpy(dict_array["phi_t"])

    def save(self, filename: str):
        dict_array = self.to_numpy()
        np.savez(filename, **dict_array)

    def load(self, filename: str):
        dict_array = np.load(filename)
        self.from_numpy(dict_array)

    def to_active_frame_set(self):
        point_np = self.point.to_numpy()
        normal_np = self.normal.to_numpy()
        phi_t_np = self.phi_t.to_numpy()

        point_np = point_np.reshape(-1, 3)
        normal_np = normal_np.reshape(-1, 2)
        phi_t_np = phi_t_np.ravel()

        is_active = ~np.isnan(point_np).any(axis=1)

        point_active = point_np[is_active]
        normal_active = normal_np[is_active]
        phi_t_active = phi_t_np[is_active]

        dict_array = {}
        dict_array["point"] = point_active
        dict_array["normal"] = normal_active
        dict_array["phi_t"] = phi_t_active

        return dict_array

    def save_active_frame_set(self, filename: str):
        dict_array = self.to_active_frame_set()
        np.savez(filename, **dict_array)

        # np.savetxt(
        #     filename,
        #     np.concatenate((point_active, orientation_active.reshape((-1, 1))), axis=1),
        #     fmt="%.3f",
        # )


class Set:
    def __init__(self):
        self.point = None
        self.normal = None
        self.phi_t = None

    def to_numpy(self):
        point_np = self.point.to_numpy()
        normal_np = self.normal.to_numpy()
        phi_t_np = self.phi_t.to_numpy()

        dict_array = {}
        dict_array["point"] = point_np
        dict_array["normal"] = normal_np
        dict_array["phi_t"] = phi_t_np

        return dict_array

    def from_numpy(self, dict_array):
        shape = dict_array["phi_t"].shape

        self.point = ti.Vector.field(n=3, dtype=ti.f32, shape=shape)
        self.normal = ti.Vector.field(n=2, dtype=ti.f32, shape=shape)
        self.phi_t = ti.field(dtype=ti.f32, shape=shape)

        self.point.from_numpy(dict_array["point"])
        self.normal.from_numpy(dict_array["normal"])
        self.phi_t.from_numpy(dict_array["phi_t"])

    def save(self, filename: str):
        dict_array = self.to_numpy()
        np.savez(filename, **dict_array)

    def load(self, filename: str):
        dict_array = np.load(filename)
        self.from_numpy(dict_array)

    def get_aabb(self):
        set_dict = self.to_numpy()
        point = set_dict["point"]
        p_min = np.min(point, axis=0)
        p_max = np.max(point, axis=0)

        return p_min, p_max

    def get_highest_point(self):
        set_dict = self.to_numpy()
        point = set_dict["point"]
        arg_max = np.argmax(point[:, 2])

        return point[arg_max]


@ti.func
def frame_set_any_point_inside_cone_i_brute_force(
    point: ti.template(),
    normal: ti.template(),
    i: int,
    cone_cos_half_angle: float,
):
    is_unaccessible_i = False
    p_i = point[i]
    n_i = direction.spherical_to_cartesian(normal[i])

    for j in range(point.shape[0]):
        p_j = point[j]
        if ti.math.isnan(p_j).any() or i == j:
            continue

        is_unaccessible_ij = cone.is_point_inside(p_j, p_i, n_i, cone_cos_half_angle)
        if is_unaccessible_ij:
            is_unaccessible_i = True

    return is_unaccessible_i