-
Notifications
You must be signed in to change notification settings - Fork 20
/
pointcnn_kitti.py
288 lines (245 loc) · 14.2 KB
/
pointcnn_kitti.py
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
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import math
import pointfly as pf
import tensorflow as tf
def xconv(pts, fts, qrs, tag, N, K, D, P, C, C_pts_fts, is_training, with_X_transformation, depth_multiplier,
sorting_method=None, with_global=False):
_, indices_dilated = pf.knn_indices_general(qrs, pts, K * D, True)
indices = indices_dilated[:, :, ::D, :]
if sorting_method is not None:
indices = pf.sort_points(pts, indices, sorting_method)
nn_pts = tf.gather_nd(pts, indices, name=tag + 'nn_pts') # (N, P, K, 3)
nn_pts_center = tf.expand_dims(qrs, axis=2, name=tag + 'nn_pts_center') # (N, P, 1, 3)
nn_pts_local = tf.subtract(nn_pts, nn_pts_center, name=tag + 'nn_pts_local') # (N, P, K, 3)
# Prepare features to be transformed
nn_fts_from_pts_0 = pf.dense(nn_pts_local, C_pts_fts, tag + 'nn_fts_from_pts_0', is_training)
nn_fts_from_pts = pf.dense(nn_fts_from_pts_0, C_pts_fts, tag + 'nn_fts_from_pts', is_training)
if fts is None:
nn_fts_input = nn_fts_from_pts
else:
nn_fts_from_prev = tf.gather_nd(fts, indices, name=tag + 'nn_fts_from_prev')
nn_fts_input = tf.concat([nn_fts_from_pts, nn_fts_from_prev], axis=-1, name=tag + 'nn_fts_input')
if with_X_transformation:
######################## X-transformation #########################
X_0 = pf.conv2d(nn_pts_local, K * K, tag + 'X_0', is_training, (1, K))
X_0_KK = tf.reshape(X_0, (N, P, K, K), name=tag + 'X_0_KK')
X_1 = pf.depthwise_conv2d(X_0_KK, K, tag + 'X_1', is_training, (1, K))
X_1_KK = tf.reshape(X_1, (N, P, K, K), name=tag + 'X_1_KK')
X_2 = pf.depthwise_conv2d(X_1_KK, K, tag + 'X_2', is_training, (1, K), activation=None)
X_2_KK = tf.reshape(X_2, (N, P, K, K), name=tag + 'X_2_KK')
fts_X = tf.matmul(X_2_KK, nn_fts_input, name=tag + 'fts_X')
###################################################################
else:
fts_X = nn_fts_input
fts_conv = pf.separable_conv2d(fts_X, C, tag + 'fts_conv', is_training, (1, K), depth_multiplier=depth_multiplier)
fts_conv_3d = tf.squeeze(fts_conv, axis=2, name=tag + 'fts_conv_3d')
if with_global:
fts_global_0 = pf.dense(qrs, C // 4, tag + 'fts_global_0', is_training, with_bn=False)
fts_global = pf.dense(fts_global_0, C // 4, tag + 'fts_global', is_training, with_bn=False)
#fts_global_0 = pf.dense(qrs, C // 4, tag + 'fts_global_0', is_training)
#fts_global = pf.dense(fts_global_0, C // 4, tag + 'fts_global', is_training)
return tf.concat([fts_global, fts_conv_3d], axis=-1, name=tag + 'fts_conv_3d_with_global')
else:
return fts_conv_3d
class PointCNN:
def __init__(self, points, features, images, image_xy, num_class, is_training, setting, task):
print("points:",points)
print("features:",features)
xconv_params = setting.xconv_params
print("xconv_params:",xconv_params)
fc_params = setting.fc_params
print("fc_params:", fc_params)
# 从配置文件中获取后续分支的参数
extra_branch_xconv_params = setting.extra_branch_xconv_params
extra_branch_fc_params = setting.extra_branch_fc_params
mask_sample_num = setting.mask_sample_num
with_X_transformation = setting.with_X_transformation
sorting_method = setting.sorting_method
N = tf.shape(points)[0]
point_num = tf.shape(points)[1]
if setting.with_fps:
from sampling import tf_sampling
self.layer_pts = [points]
if features is None:
self.layer_fts = [features]
else:
C_fts = xconv_params[0][-1] // 2
features_hd = pf.dense(features, C_fts, 'features_hd', is_training)
self.layer_fts = [features_hd]
for layer_idx, layer_param in enumerate(xconv_params):
tag = 'xconv_' + str(layer_idx + 1) + '_'
K, D, P, C = layer_param
fts = self.layer_fts[-1]
# get k centroid points
pts = self.layer_pts[-1]
if P == -1:
qrs = self.layer_pts[-1]
else:
if setting.with_fps:
fps_indices = tf_sampling.farthest_point_sample(P, pts)
batch_indices = tf.tile(tf.reshape(tf.range(N), (-1, 1, 1)), (1, P, 1))
fps_indices_g = tf.concat([batch_indices, tf.expand_dims(fps_indices,-1)], axis=-1)
qrs = tf.gather_nd(pts, fps_indices_g, name= tag + 'qrs') # (N, P, 3)
else:
qrs = tf.slice(pts, (0, 0, 0), (-1, P, -1), name=tag + 'qrs') # (N, P, 3)
self.layer_pts.append(qrs)
if layer_idx == 0:
C_pts_fts = C // 2 if fts is None else C // 4
depth_multiplier = 4
else:
C_prev = xconv_params[layer_idx - 1][-1]
C_pts_fts = C_prev // 4
depth_multiplier = math.ceil(C / C_prev)
fts_xconv = xconv(pts, fts, qrs, tag, N, K, D, P, C, C_pts_fts, is_training, with_X_transformation,
depth_multiplier, sorting_method,False)
self.layer_fts.append(fts_xconv)
if task == 'segmentation':
for layer_idx, layer_param in enumerate(setting.xdconv_params):
tag = 'xdconv_' + str(layer_idx + 1) + '_'
K, D, pts_layer_idx, qrs_layer_idx = layer_param
pts = self.layer_pts[pts_layer_idx + 1]
fts = self.layer_fts[pts_layer_idx + 1] if layer_idx == 0 else self.layer_fts[-1]
qrs = self.layer_pts[qrs_layer_idx + 1]
fts_qrs = self.layer_fts[qrs_layer_idx + 1]
_, _, P, C = xconv_params[qrs_layer_idx]
C_pts_fts = C_prev // 4
depth_multiplier = 1
fts_xdconv = xconv(pts, fts, qrs, tag, N, K, D, P, C, C_pts_fts, is_training, with_X_transformation,
depth_multiplier, sorting_method)
fts_concat = tf.concat([fts_xdconv, fts_qrs], axis=-1, name=tag + 'fts_concat')
fts_fuse = pf.dense(fts_concat, C, tag + 'fts_fuse', is_training)
self.layer_pts.append(qrs)
self.layer_fts.append(fts_fuse)
self.fc_layers = [self.layer_fts[-1]]
for layer_idx, layer_param in enumerate(fc_params):
channel_num, drop_rate = layer_param
fc = pf.dense(self.fc_layers[-1], channel_num, 'fc{:d}'.format(layer_idx), is_training)
fc_drop = tf.layers.dropout(fc, drop_rate, training=is_training, name='fc{:d}_drop'.format(layer_idx))
self.fc_layers.append(fc_drop)
if task == 'classification':
fc_mean = tf.reduce_mean(self.fc_layers[-1], axis=1, keep_dims=True, name='fc_mean')
self.fc_layers[-1] = tf.cond(is_training, lambda: self.fc_layers[-1], lambda: fc_mean)
self.logits = pf.dense(self.fc_layers[-1], num_class, 'logits', is_training, with_bn=False, activation=None)
self.probs = tf.nn.softmax(self.logits, name='probs')
# 从分割结果获取Mask点云(车辆点云)
probs_others = tf.slice(self.probs, [0, 0, 0], [N, point_num, 1])
probs_instance = tf.slice(self.probs, [0, 0, 1], [N, point_num, 1])
hold_sub_mask = tf.subtract(probs_instance, probs_others, name="hold_sub_mask")
indices_mask = tf.py_func(pf.instance_choice, [hold_sub_mask, mask_sample_num], tf.int32)
indices_mask.set_shape([hold_sub_mask.get_shape()[0], mask_sample_num, 2])
points_mask = tf.gather_nd(points, indices_mask, name='points_mask')
# 计算mask点云重心
if setting.with_xyz_res:
points_mask_mean = tf.reduce_mean(points_mask, axis=-2, keep_dims=True,name='points_mask_mean')
points_mask_mean_sq = tf.squeeze(points_mask_mean, name='points_mask_mean_sq')
points_mask_centroid = tf.subtract(points_mask, points_mask_mean, name='points_mask_centroid')
eb_pts = points_mask_centroid
else:
eb_pts = points_mask
# 获取mask点云特征
if features is None:
features_mask_hd = None
else:
C_fts = extra_branch_xconv_params[0][-1] // 2
features_mask = tf.gather_nd(features, indices_mask, name='features_mask')
mask_features_hd = pf.dense(features_mask, C_fts, 'mask_features_hd', is_training)
eb_fts = mask_features_hd
# Bbox分支(尺寸与位置分支)
self.bbox_layer_fts = [eb_fts]
self.bbox_layer_pts = [eb_pts]
for layer_idx, layer_param in enumerate(extra_branch_xconv_params):
tag = 'bbox_branch_xconv_' + str(layer_idx + 1) + '_'
K, D, P, C = layer_param
pts = self.bbox_layer_pts[-1]
fts = self.bbox_layer_fts[-1]
# get centroid points
if P == -1:
qrs = eb_pts
else:
if setting.eb_with_fps:
fps_indices = tf_sampling.farthest_point_sample(P, pts)
batch_indices = tf.tile(tf.reshape(tf.range(N), (-1, 1, 1)), (1, P, 1))
fps_indices_g = tf.concat([batch_indices, tf.expand_dims(fps_indices, -1)], axis=-1)
qrs = tf.gather_nd(pts, fps_indices_g, name=tag + 'qrs') # (N, P, 3)
else:
qrs = tf.slice(pts, (0, 0, 0), (-1, P, -1), name=tag + 'qrs') # (N, P, 3)
self.bbox_layer_pts.append(qrs)
if layer_idx == 0:
C_pts_fts = C // 2 if fts is None else C // 4
depth_multiplier = 4
else:
C_prev = xconv_params[layer_idx - 1][-1]
C_pts_fts = C_prev // 4
depth_multiplier = math.ceil(C / C_prev)
fts_xconv = xconv(pts, fts, qrs, tag, N, K, D, P, C, C_pts_fts, is_training,
with_X_transformation, depth_multiplier, sorting_method, False)
self.bbox_layer_fts.append(fts_xconv)
# HWL 预测分支 以用maxpooling特征进行预测为例
self.hwl_fc_layers = [tf.reduce_max(self.bbox_layer_fts[-1], axis=1, name='hwl_fts_max')]
for layer_idx, layer_param in enumerate(extra_branch_fc_params):
channel_num, drop_rate = layer_param
fc = pf.dense(self.hwl_fc_layers[-1], channel_num, 'hwl_fc{:d}'.format(layer_idx),
is_training, with_bn=setting.hwl_branch_setting["with_bn"])
fc_drop = tf.layers.dropout(fc, drop_rate, training=is_training,
name='hwl_fc{:d}_drop'.format(layer_idx))
self.hwl_fc_layers.append(fc_drop)
self.logits_hwl = pf.dense(self.hwl_fc_layers[-1], 3, 'logits_hwl', is_training, with_bn=False,
activation=None)
# XYZ 预测分支 以用maxpooling特征进行预测为例
self.xyz_fc_layers = [tf.reduce_max(self.bbox_layer_fts[-1], axis=1, name='xyz_fts_max')]
for layer_idx, layer_param in enumerate(extra_branch_fc_params):
channel_num, drop_rate = layer_param
fc = pf.dense(self.xyz_fc_layers[-1], channel_num, 'xyz_fc{:d}'.format(layer_idx),
is_training, with_bn=setting.xyz_branch_setting["with_bn"])
fc_drop = tf.layers.dropout(fc, drop_rate, training=is_training,
name='xyz_fc{:d}_drop'.format(layer_idx))
self.xyz_fc_layers.append(fc_drop)
if setting.with_xyz_res:
logits_xyz_res = pf.dense(self.xyz_fc_layers[-1], 3, 'logits_xyz_res', is_training,
with_bn=False, activation=None)
self.logits_xyz = tf.add(points_mask_mean_sq, logits_xyz_res, name='logits_xyz')
else:
self.logits_xyz = pf.dense(self.xyz_fc_layers[-1], 3, 'logits_xyz_res', is_training,
with_bn=False, activation=None)
# Ry分支(旋转角分支)
self.ry_layer_fts = [eb_fts]
self.ry_layer_pts = [eb_pts]
for layer_idx, layer_param in enumerate(extra_branch_xconv_params):
tag = 'ry_branch_xconv_' + str(layer_idx + 1) + '_'
K, D, P, C = layer_param
fts = self.ry_layer_fts[-1]
# get centroid points
pts = self.ry_layer_pts[-1]
if P == -1:
qrs = eb_pts
else:
if setting.eb_with_fps:
fps_indices = tf_sampling.farthest_point_sample(P, pts)
qrs = tf_sampling.gather_point(pts, fps_indices) # (N,P,3)
else:
qrs = tf.slice(pts, (0, 0, 0), (-1, P, -1), name=tag + 'qrs') # (N, P, 3)
self.ry_layer_pts.append(qrs)
if layer_idx == 0:
C_pts_fts = C // 2 if fts is None else C // 4
depth_multiplier = 4
else:
C_prev = xconv_params[layer_idx - 1][-1]
C_pts_fts = C_prev // 4
depth_multiplier = math.ceil(C / C_prev)
fts_xconv = xconv(pts, fts, qrs, tag, N, K, D, P, C, C_pts_fts, is_training,
with_X_transformation, depth_multiplier, sorting_method, False)
self.ry_layer_fts.append(fts_xconv)
# Ry 分类预测分支 以用maxpooling特征进行预测为例
self.ry_fc_layers = [tf.reduce_max(self.ry_layer_fts[-1], axis=1, name='ry_fts_max')]
for layer_idx, layer_param in enumerate(extra_branch_fc_params):
channel_num, drop_rate = layer_param
fc = pf.dense(self.ry_fc_layers[-1], channel_num, 'ry_fc{:d}'.format(layer_idx),
is_training, with_bn=setting.ry_branch_setting["with_bn"])
fc_drop = tf.layers.dropout(fc, drop_rate, training=is_training,
name='ry_fc{:d}_drop'.format(layer_idx))
self.ry_fc_layers.append(fc_drop)
self.logits_ry = pf.dense(self.ry_fc_layers[-1], 24, 'logits_ry', is_training,
with_bn=False, activation=None)
self.probs_ry = tf.nn.softmax(self.logits_ry, name='probs_ry')