dmytromishkin
commited on
Commit
•
420d591
1
Parent(s):
b513ce3
accept w/o semantics
Browse files- hoho/hoho.py +11 -18
- hoho/vis.py +8 -1
hoho/hoho.py
CHANGED
@@ -4,6 +4,14 @@ import shutil
|
|
4 |
from pathlib import Path
|
5 |
from typing import Dict
|
6 |
import warnings
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
|
8 |
from PIL import ImageFile
|
9 |
|
@@ -40,13 +48,7 @@ def setup(local_dir='./data/usm-training-data/data'):
|
|
40 |
LOCAL_DATADIR.mkdir(parents=True)
|
41 |
|
42 |
return LOCAL_DATADIR
|
43 |
-
|
44 |
-
|
45 |
-
|
46 |
-
|
47 |
-
import importlib
|
48 |
-
from pathlib import Path
|
49 |
-
import subprocess
|
50 |
|
51 |
def download_package(package_name, path_to_save='packages'):
|
52 |
"""
|
@@ -139,9 +141,7 @@ def Rt_to_eye_target(im, K, R, t):
|
|
139 |
|
140 |
|
141 |
########## general utilities ##########
|
142 |
-
|
143 |
-
import tempfile
|
144 |
-
from pathlib import Path
|
145 |
|
146 |
@contextlib.contextmanager
|
147 |
def working_directory(path):
|
@@ -184,10 +184,6 @@ def proc(row, split='train'):
|
|
184 |
return Sample(out)
|
185 |
|
186 |
|
187 |
-
|
188 |
-
|
189 |
-
|
190 |
-
|
191 |
from . import read_write_colmap
|
192 |
def decode_colmap(s):
|
193 |
with temp_working_directory():
|
@@ -209,8 +205,7 @@ def decode_colmap(s):
|
|
209 |
)
|
210 |
return cameras, images, points3D
|
211 |
|
212 |
-
|
213 |
-
import io
|
214 |
def decode(row):
|
215 |
cameras, images, points3D = decode_colmap(row)
|
216 |
|
@@ -288,8 +283,6 @@ def get_params():
|
|
288 |
|
289 |
|
290 |
|
291 |
-
import webdataset as wds
|
292 |
-
import numpy as np
|
293 |
|
294 |
|
295 |
SHARD_IDS = {'train': (0, 25), 'val': (25, 26), 'public': (26, 27), 'private': (27, 32)}
|
|
|
4 |
from pathlib import Path
|
5 |
from typing import Dict
|
6 |
import warnings
|
7 |
+
import contextlib
|
8 |
+
import tempfile
|
9 |
+
from PIL import Image
|
10 |
+
import io
|
11 |
+
import webdataset as wds
|
12 |
+
import numpy as np
|
13 |
+
import importlib
|
14 |
+
import subprocess
|
15 |
|
16 |
from PIL import ImageFile
|
17 |
|
|
|
48 |
LOCAL_DATADIR.mkdir(parents=True)
|
49 |
|
50 |
return LOCAL_DATADIR
|
51 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
52 |
|
53 |
def download_package(package_name, path_to_save='packages'):
|
54 |
"""
|
|
|
141 |
|
142 |
|
143 |
########## general utilities ##########
|
144 |
+
|
|
|
|
|
145 |
|
146 |
@contextlib.contextmanager
|
147 |
def working_directory(path):
|
|
|
184 |
return Sample(out)
|
185 |
|
186 |
|
|
|
|
|
|
|
|
|
187 |
from . import read_write_colmap
|
188 |
def decode_colmap(s):
|
189 |
with temp_working_directory():
|
|
|
205 |
)
|
206 |
return cameras, images, points3D
|
207 |
|
208 |
+
|
|
|
209 |
def decode(row):
|
210 |
cameras, images, points3D = decode_colmap(row)
|
211 |
|
|
|
283 |
|
284 |
|
285 |
|
|
|
|
|
286 |
|
287 |
|
288 |
SHARD_IDS = {'train': (0, 25), 'val': (25, 26), 'public': (26, 27), 'private': (27, 32)}
|
hoho/vis.py
CHANGED
@@ -51,7 +51,14 @@ def show_wf(row, radius=10):
|
|
51 |
'valley',
|
52 |
'hip',
|
53 |
'transition_line']
|
54 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
55 |
# return [line(a,b, radius=radius, c=color_mappings.edge_colors[cls_id]) for (a,b), cls_id in zip(np.stack([*row['wf_vertices']])[np.stack(row['wf_edges'])], row['edge_semantics'])]
|
56 |
|
57 |
|
|
|
51 |
'valley',
|
52 |
'hip',
|
53 |
'transition_line']
|
54 |
+
if 'edge_semantics' not in row:
|
55 |
+
print ("Warning: edge semantics is not here, skipping")
|
56 |
+
return [line(a,b, radius=radius, c=(214, 251, 248)) for a,b in np.stack([*row['wf_vertices']])[np.stack(row['wf_edges'])]]
|
57 |
+
elif len(np.stack(row['wf_edges'])) == len(row['edge_semantics']):
|
58 |
+
return [line(a,b, radius=radius, c=color_mappings.gestalt_color_mapping[EDGE_CLASSES[cls_id]]) for (a,b), cls_id in zip(np.stack([*row['wf_vertices']])[np.stack(row['wf_edges'])], row['edge_semantics'])]
|
59 |
+
else:
|
60 |
+
print ("Warning: edge semantics has different length compared to edges, skipping semantics")
|
61 |
+
return [line(a,b, radius=radius, c=(214, 251, 248)) for a,b in np.stack([*row['wf_vertices']])[np.stack(row['wf_edges'])]]
|
62 |
# return [line(a,b, radius=radius, c=color_mappings.edge_colors[cls_id]) for (a,b), cls_id in zip(np.stack([*row['wf_vertices']])[np.stack(row['wf_edges'])], row['edge_semantics'])]
|
63 |
|
64 |
|