UVStyle-Net/graph_plotter.py at master · Fihven/UVStyle-Net · 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
import numpy as np
import pandas as pd
import plotly.graph_objs as go
import streamlit as st
import trimesh
from sklearn.preprocessing import StandardScaler

from datasets.solid_letters import SolidLETTERSSubset


def face_adjacency_plot(graph, node_colors=None):
    x = graph.ndata['x']
    df = pd.DataFrame(x[:, :, :, :3].flatten(end_dim=-2).numpy(), columns=['x', 'y', 'z'])
    df[['normal_x', 'normal_y', 'normal_z']] = pd.DataFrame(x[:, :, :, 3:6].flatten(end_dim=-2).numpy())
    df['mask'] = x[:, :, :, 6].reshape([-1, 1]).numpy()

    df['node'] = np.array([[i] * 100 for i in range(graph.number_of_nodes())]).flatten()

    masked = df[df['mask'] == 1.]

    # face adjacency graph
    face_adj = masked.groupby('node').mean()
    if node_colors is None:
        node_colors = face_adj.index
    src, dst, _ = zip(*list(graph.to_networkx().edges))
    src = face_adj.loc[list(src)]
    dst = face_adj.loc[list(dst)]
    face_adj_fig = go.Figure(data=go.Scatter3d(x=face_adj['x'],
                                               y=face_adj['y'],
                                               z=face_adj['z'],
                                               marker=dict(
                                                   color=node_colors
                                               ),
                                               mode='markers'))
    for i in range(len(src)):
        face_adj_fig.add_scatter3d(
            # Line reference to the axes
            x=[src.iloc[i]['x'], dst.iloc[i]['x']],
            y=[src.iloc[i]['y'], dst.iloc[i]['y']],
            z=[src.iloc[i]['z'], dst.iloc[i]['z']],
            mode='lines',
            line=dict(
                color="Black",
                width=1,
            )
        )
    face_adj_fig.update_layout({
        'showlegend': False
    })
    return face_adj_fig


def graph_to_xyz_mask(graph):
    x = graph.ndata['x']
    xyz = x[:, :, :, :3].reshape([-1, 3])
    mask = x[:, :, :, 6].reshape([-1])
    return xyz, mask


def uv_samples_plot(xyz, mask,
                    xyz_grads=None,
                    scale_xyz_grads=0.05,
                    marker_size=3,
                    mesh: trimesh.Trimesh = None,
                    mesh_alpha=1.,
                    corner_align=False):

    if mask is not None:
        xyz = xyz[mask == 1]

    scatter = go.Scatter3d(x=xyz[:, 0],
                           y=xyz[:, 1],
                           z=xyz[:, 2],
                           marker=dict(
                               size=marker_size,
                               color='rgb(0, 0, 255)'
                           ),

                           mode='markers')
    fig = go.Figure(data=scatter)

    if mesh is not None:
        if corner_align:
            mesh.vertices -= mesh.bounding_box.bounds[0]
        mesh_plot = go.Mesh3d(x=mesh.vertices[:, 0],
                              y=mesh.vertices[:, 1],
                              z=mesh.vertices[:, 2],
                              i=mesh.faces[:, 0],
                              j=mesh.faces[:, 1],
                              k=mesh.faces[:, 2],
                              color='gray',
                              opacity=mesh_alpha)
        fig.add_trace(mesh_plot)

    if xyz_grads is not None:
        if mask is not None:
            xyz_grads = xyz_grads[mask == 1]
        xyz_grads = StandardScaler(with_mean=False).fit_transform(xyz_grads) * scale_xyz_grads
        # hist = ff.create_distplot([xyz_grads[:, 0].tolist(), xyz_grads[:, 1].tolist(), xyz_grads[:, 2].tolist()], bin_size=0.5, group_labels=['x', 'y', 'z'])
        # st.plotly_chart(go.Figure(hist))
        for i in range(len(xyz)):
            x, y, z = xyz[i]
            x_, y_, z_ = xyz_grads[i]
            fig.add_scatter3d(
                x=[x, x + x_],
                y=[y, y + y_],
                z=[z, z + z_],
                mode='lines',
                line=dict(
                    color="Black",
                    width=3,
                )
            )
    fig.update_layout({
        'showlegend': False,
        'scene': {
            'xaxis': {
                'zeroline': False,
                'showline': False,
                'showgrid': False,
                'showbackground': False,
                'visible': False
            },
            'yaxis': {
                'zeroline': False,
                'showline': False,
                'showgrid': False,
                'showbackground': False,
                'visible': False
            },
            'zaxis': {
                'zeroline': False,
                'showline': False,
                'showgrid': False,
                'showbackground': False,
                'visible': False
            }
        }
    })
    return fig


if __name__ == '__main__':
    dset = SolidLETTERSSubset(root_dir='dataset/bin', split='test')
    for i in range(5):
        graph, label, meta, image, graph_file = dset[i]

        st.plotly_chart(face_adjacency_plot(graph))

        st.plotly_chart(uv_samples_plot(*graph_to_xyz_mask(graph)))