def evaluate_masks(
json_dataset,
all_boxes,
all_segms,
output_dir,
use_salt=True,
cleanup=False
):
if cfg.CLUSTER.ON_CLUSTER:
# On the cluster avoid saving these files in the job directory
output_dir = '/tmp'
res_file = os.path.join(
output_dir, 'segmentations_' + json_dataset.name + '_results')
if use_salt:
res_file += '_{}'.format(str(uuid.uuid4()))
res_file += '.json'
results_dir = os.path.join(output_dir, 'results')
if not os.path.exists(results_dir):
os.mkdir(results_dir)
os.environ['CITYSCAPES_DATASET'] = DATASETS[json_dataset.name][RAW_DIR]
os.environ['CITYSCAPES_RESULTS'] = output_dir
# Load the Cityscapes eval script *after* setting the required env vars,
# since the script reads their values into global variables (at load time).
import cityscapesscripts.evaluation.evalInstanceLevelSemanticLabeling \
as cityscapes_eval
roidb = json_dataset.get_roidb()
for i, entry in enumerate(roidb):
im_name = entry['image']
basename = os.path.splitext(os.path.basename(im_name))[0]
txtname = os.path.join(output_dir, basename + 'pred.txt')
with open(txtname, 'w') as fid_txt:
if i % 10 == 0:
logger.info('i: {}: {}'.format(i, basename))
for j in range(1, len(all_segms)):
clss = json_dataset.classes[j]
clss_id = cityscapes_eval.name2label[clss].id
segms = all_segms[j][i]
boxes = all_boxes[j][i]
if segms == []:
continue
masks = mask_util.decode(segms)
for k in range(boxes.shape[0]):
score = boxes[k, -1]
mask = masks[:, :, k]
pngname = os.path.join(
'results',
basename + '_' + clss + '_{}.png'.format(k))
# write txt
fid_txt.write('{} {} {}\n'.format(pngname, clss_id, score))
# save mask
cv2.imwrite(os.path.join(output_dir, pngname), mask * 255)
logger.info('Evaluating...')
cityscapes_eval.main([])
return None
def evaluate_masks(
json_dataset,
all_boxes,
all_segms,
output_dir,
use_salt=True,
cleanup=False
):
if cfg.CLUSTER.ON_CLUSTER:
# On the cluster avoid saving these files in the job directory
output_dir = '/tmp'
res_file = os.path.join(
output_dir, 'segmentations_' + json_dataset.name + '_results')
if use_salt:
res_file += '_{}'.format(str(uuid.uuid4()))
res_file += '.json'
results_dir = os.path.join(output_dir, 'results')
if not os.path.exists(results_dir):
os.mkdir(results_dir)
os.environ['CITYSCAPES_DATASET'] = get_raw_dir(json_dataset.name)
os.environ['CITYSCAPES_RESULTS'] = output_dir
# Load the Cityscapes eval script *after* setting the required env vars,
# since the script reads their values into global variables (at load time).
import cityscapesscripts.evaluation.evalInstanceLevelSemanticLabeling \
as cityscapes_eval
roidb = json_dataset.get_roidb()
for i, entry in enumerate(roidb):
im_name = entry['image']
basename = os.path.splitext(os.path.basename(im_name))[0]
txtname = os.path.join(output_dir, basename + 'pred.txt')
with open(txtname, 'w') as fid_txt:
if i % 10 == 0:
logger.info('i: {}: {}'.format(i, basename))
for j in range(1, len(all_segms)):
clss = json_dataset.classes[j]
clss_id = cityscapes_eval.name2label[clss].id
segms = all_segms[j][i]
boxes = all_boxes[j][i]
if segms == []:
continue
masks = mask_util.decode(segms)
for k in range(boxes.shape[0]):
score = boxes[k, -1]
mask = masks[:, :, k]
pngname = os.path.join(
'results',
basename + '_' + clss + '_{}.png'.format(k))
# write txt
fid_txt.write('{} {} {}\n'.format(pngname, clss_id, score))
# save mask
cv2.imwrite(os.path.join(output_dir, pngname), mask * 255)
logger.info('Evaluating...')
cityscapes_eval.main([])
return None
def evaluate_masks(
self,
all_boxes,
all_segms,
output_dir,
):
res_file = os.path.join(
output_dir, 'segmentations_' + self.dataset.name + '_results')
res_file += '.json'
#os.environ['CITYSCAPES_DATASET'] = os.path.join(os.path.dirname(__file__), '../../data/cityscapes')
os.environ['CITYSCAPES_DATASET'] = '/n/pfister_lab2/Lab/xingyu/Dataset/Cityscapes'
os.environ['CITYSCAPES_RESULTS'] = os.path.join(output_dir, 'inst_seg')
sys.path.insert(0, os.path.join(os.path.abspath(os.path.dirname(__file__)), '..', '..', 'lib', 'dataset_devkit', 'cityscapesScripts'))
sys.path.insert(0, os.path.join(os.path.abspath(os.path.dirname(__file__)), '..', '..', 'lib', 'dataset_devkit', 'cityscapesScripts', 'cityscapesscripts', 'evaluation'))
# Load the Cityscapes eval script *after* setting the required env vars,
# since the script reads their values into global variables (at load time).
import cityscapesscripts.evaluation.evalInstanceLevelSemanticLabeling \
as cityscapes_eval
sys.argv = []
roidb = self.dataset.get_roidb()
for i, entry in enumerate(roidb):
im_name = entry['image']
basename = os.path.splitext(os.path.basename(im_name))[0]
txtname = os.path.join(output_dir, 'inst_seg', basename + 'pred.txt')
os.makedirs(os.path.join(output_dir, 'inst_seg'), exist_ok=True)
with open(txtname, 'w') as fid_txt:
for j in range(1, len(all_segms)):
clss = self.dataset.classes[j]
clss_id = cityscapes_eval.name2label[clss].id
segms = all_segms[j][i]
boxes = all_boxes[j][i]
if segms == []:
continue
masks = mask_util.decode(segms)
for k in range(boxes.shape[0]):
score = boxes[k, -1]
mask = masks[:, :, k]
pngname = os.path.join(
'seg_results', basename,
basename + '_' + clss + '_{}.png'.format(k))
# write txt
fid_txt.write('{} {} {}\n'.format(pngname, clss_id, score))
# save mask
os.makedirs(os.path.join(output_dir, 'inst_seg', 'seg_results', basename), exist_ok=True)
cv2.imwrite(os.path.join(output_dir, 'inst_seg', pngname), mask * 255)
cityscapes_eval.main()
return None
def evaluate_instance_segmentations(
json_dataset_name, inp_results_dir, out_results_dir):
if cfg.CLUSTER.ON_CLUSTER:
# On the cluster avoid saving these files in the job directory
inp_results_dir = '/tmp'
res_file = os.path.join(
inp_results_dir, 'segmentations_' + json_dataset_name + '_results')
res_file += '.json'
results_dir = os.path.join(inp_results_dir, 'results')
if not os.path.exists(results_dir):
os.mkdir(results_dir)
os.environ['CITYSCAPES_RESULTS'] = inp_results_dir
os.environ['CITYSCAPES_DATASET'] = get_raw_dir(json_dataset_name)
os.environ['OUTPUTS_DIR'] = out_results_dir
import cityscapesscripts.evaluation.evalInstanceLevelSemanticLabeling as cityscapes_eval
logger.info('Evaluating...')
cityscapes_eval.main()
def evaluate_masks(json_dataset,
all_boxes,
all_segms,
output_dir,
use_salt=True,
cleanup=False):
if cfg.CLUSTER.ON_CLUSTER:
# On the cluster avoid saving these files in the job directory
output_dir = '/tmp'
res_file = os.path.join(output_dir,
'segmentations_' + json_dataset.name + '_results')
if use_salt:
res_file += '_{}'.format(str(uuid.uuid4()))
res_file += '.json'
results_dir = os.path.join(output_dir, 'results')
if not os.path.exists(results_dir):
os.mkdir(results_dir)
os.environ['CITYSCAPES_DATASET'] = DATASETS[json_dataset.name][RAW_DIR]
os.environ['CITYSCAPES_RESULTS'] = output_dir
# Load the Cityscapes eval script *after* setting the required env vars,
# since the script reads their values into global variables (at load time).
import cityscapesscripts.evaluation.evalInstanceLevelSemanticLabeling \
as cityscapes_eval
if not glob.glob(cityscapes_eval.args.groundTruthSearch):
logger.warning(
"Skipping maks test since groundtruths not found by pattern " +
cityscapes_eval.args.groundTruthSearch)
return None
cityscapes_eval.args.gtInstancesFile = os.path.abspath(
os.path.join(
DATASETS[json_dataset.name][RAW_DIR],
"../cityscapesscripts-evaluation-cached-gtInstances.json"))
roidb = json_dataset.get_roidb()
for i, entry in enumerate(roidb):
im_name = entry['image']
basename = os.path.splitext(os.path.basename(im_name))[0]
txtname = os.path.join(output_dir, basename + 'pred.txt')
with open(txtname, 'w') as fid_txt:
if i % 10 == 0:
logger.info('i: {}: {}'.format(i, basename))
for j in range(1, len(all_segms)):
clss = json_dataset.classes[j]
clss_id = cityscapes_eval.name2label[clss].id
segms = all_segms[j][i]
boxes = all_boxes[j][i]
if segms == []:
continue
masks = mask_util.decode(segms)
for k in range(boxes.shape[0]):
score = boxes[k, -1]
mask = masks[:, :, k]
pngname = os.path.join(
'results', basename + '_' + clss + '_{}.png'.format(k))
# write txt
fid_txt.write('{} {} {}\n'.format(pngname, clss_id, score))
# save mask
cv2.imwrite(os.path.join(output_dir, pngname), mask * 255)
logger.info('Evaluating...')
argv_old = sys.argv
sys.argv = ["force_only_unused_arg_0_workaround_for_cityscapes_eval"]
cityscapes_eval.main()
sys.argv = argv_old
return None
# Instance Stixels is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# Instance Stixels is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Instance Stixels. If not, see <http://www.gnu.org/licenses/>.
import os
import sys
import argparse
import cityscapesscripts.evaluation.evalInstanceLevelSemanticLabeling as cs
if __name__ == "__main__":
# This is soooo ugly, but it works.
gt_path = sys.argv[1]
results_path = sys.argv[2]
sys.argv = []
cs.args.groundTruthSearch =\
os.path.join(os.path.abspath(gt_path), "*_gtFine_instanceIds.png")
cs.args.predictionPath = os.path.abspath(results_path)
print(cs.args.groundTruthSearch)
print(cs.args.predictionPath)
cs.main()
