default='../output/results/detections')
args = parser.parse_args()
return args
if __name__ == '__main__':
args = parse_args()
if not os.path.isdir(args.out_dir):
os.makedirs(args.out_dir)
# Load the nucoco dataset
ann_file = args.ann_file
img_dir = args.imgs_dir
coco = COCO_PLUS(ann_file, img_dir)
#img_ids=[10000323,10000193]
img_ids = [10009844, 10026493]
# print(coco.imgs)
# input('something')
for img_id in img_ids:
print(img_id)
#fig, ax = plt.subplots( nrows=1, ncols=1 )
points = coco.imgToPointcloud[img_id]['points']
ann_ids = coco.getAnnIds(img_id)
anns = coco.loadAnns(ann_ids)
# bboxes = anns['bbox']
img_path = os.path.join(img_dir, coco.imgs[img_id]["file_name"])
img = cv2.imread(img_path)
def main(args):
logger = logging.getLogger(__name__)
dummy_coco_dataset = dummy_datasets.get_coco_dataset()
cfg_orig = load_cfg(envu.yaml_dump(cfg))
proposals = rrpn_loader(args.rpn_pkl)
coco = COCO_PLUS(args.ann_file, args.imgs_dir)
image_id = coco.dataset['images'][36]['id']
img_path = os.path.join(args.imgs_dir, coco.imgs[image_id]["file_name"])
if args.im_file is None:
im = cv2.imread(img_path)
args.im_file = img_path
else:
im = cv2.imread(args.im_file)
# img2 = Image.fromarray(im[:,:,::-1], 'RGB')
# img2.show()
# input('something')
## Get the proposals for this image
proposal_boxes = proposals[image_id]['boxes']
_proposal_scores = proposals[image_id]['scores']
workspace.ResetWorkspace()
# print(proposal_boxes)
# input('something')
cls_boxes, cls_segms, cls_keyps = None, None, None
for i in range(0, len(args.models_to_run), 2):
pkl = args.models_to_run[i]
yml = args.models_to_run[i + 1]
cfg.immutable(False)
merge_cfg_from_cfg(cfg_orig)
merge_cfg_from_file(yml)
if len(pkl) > 0:
weights_file = pkl
else:
weights_file = cfg.TEST.WEIGHTS
cfg.NUM_GPUS = 1
assert_and_infer_cfg(cache_urls=False)
model = model_engine.initialize_model_from_cfg(weights_file)
with c2_utils.NamedCudaScope(0):
cls_boxes_, cls_segms_, cls_keyps_ = \
model_engine.im_detect_all(model, im, proposal_boxes)
cls_boxes = cls_boxes_ if cls_boxes_ is not None else cls_boxes
cls_segms = cls_segms_ if cls_segms_ is not None else cls_segms
cls_keyps = cls_keyps_ if cls_keyps_ is not None else cls_keyps
workspace.ResetWorkspace()
out_name = os.path.join(
args.output_dir, '{}'.format(os.path.basename(args.im_file) + '.pdf'))
logger.info('Processing {} -> {}'.format(args.im_file, out_name))
vis_utils.vis_one_image(im[:, :, ::-1],
args.im_file,
args.output_dir,
cls_boxes,
cls_segms,
cls_keyps,
dataset=dummy_coco_dataset,
box_alpha=0.3,
show_class=True,
thresh=0.7,
kp_thresh=2)
import _init_paths
import cv2
import os
import sys
import matplotlib
# matplotlib.use('Agg')
import matplotlib.pyplot as plt
from PIL import Image
from pprint import pprint
from pycocotools_plus.coco import COCO_PLUS
coco_ann_file = '/ssd_scratch/mrnabati/occlusionDetection/Code/data/coco/annotations/instances_train2017.json'
coco_imgs_dir = '/ssd_scratch/mrnabati/occlusionDetection/Code/data/coco/images/train2017'
coco = COCO_PLUS(coco_ann_file, coco_imgs_dir)
count = 0
for key, val in coco.cats.items():
count += 1
print('{}:{}'.format(key, val['name']))
print(count)
for image in coco.dataset['images']:
img_id = image['id']
ann_ids = coco.getAnnIds([img_id])
anns = coco.loadAnns(ann_ids)
print(anns[0])
print(anns[0]['bbox'][0].dtype)
input('something')
assert args.train_ratio >= 0 and args.train_ratio <= 1, \
"--train_ratio must be in range [0 1]"
assert args.include_sweeps in ['True','False'], \
"--include_sweeps must be 'True' or 'False'"
return args
#-------------------------------------------------------------------------------
if __name__ == '__main__':
random.seed(13)
args = parse_args()
nusc = NuScenes(version='v0.1', dataroot=args.dataroot, verbose=True)
coco_train = COCO_PLUS(args.train_ann_file,
args.train_imgs_dir,
new_dataset=True)
coco_val = COCO_PLUS(args.val_ann_file,
args.val_imgs_dir,
new_dataset=True)
for i in tqdm(range(0, len(nusc.scene))):
scene = nusc.scene[i]
scene_rec = nusc.get('scene', scene['token'])
sample_rec = nusc.get('sample', scene_rec['first_sample_token'])
## Get front sensors data
f_cam_rec = nusc.get('sample_data', sample_rec['data']['CAM_FRONT'])
f_rad_rec = nusc.get('sample_data', sample_rec['data']['RADAR_FRONT'])
## Get rear sensors data
default=0)
args = parser.parse_args()
return args
if __name__ == '__main__':
args = parse_args()
output_file = args.output_file
boxes = []
scores = []
ids = []
img_ind = 0
## Load the nucoco dataset
coco = COCO_PLUS(args.ann_file, args.imgs_dir)
for img_id, img_info in tqdm(coco.imgs.items()):
img_ind += 1
if int(args.include_depth)==1:
proposals = np.empty((0,5), np.float32)
else:
proposals = np.empty((0,4), np.float32)
img_width = img_info['width']
img_height = img_info['height']
pointcloud = coco.imgToPointcloud[img_id]
# pointcloud = coco.pointclouds[pc_id]
help='Dataset name according to dataset_catalog',
default='nucoco_train')
args = parser.parse_args()
return args
if __name__ == '__main__':
args = parse_args()
# fig = plt.figure(figsize=(16, 6))
# Load the nucoco dataset
dataset_name = args.dataset_name
ann_file = dataset_catalog.get_ann_fn(dataset_name)
img_dir = dataset_catalog.get_im_dir(dataset_name)
coco = COCO_PLUS(ann_file, img_dir)
# Load the proposals
proposals = rrpn_loader(args.proposals_file)
for i in range(1, len(coco.dataset['images']), 10):
fig = plt.figure(figsize=(16, 6))
img_id = coco.dataset['images'][i]['id']
scores = proposals[img_id]['scores']
boxes = proposals[img_id]['boxes']
points = coco.imgToPointcloud[img_id]['points']
img_path = os.path.join(img_dir, coco.imgs[img_id]["file_name"])
# print(img_path)
img = np.array(plt.imread(img_path))
parser = argparse.ArgumentParser(description='Explore NuCOCO')
parser.add_argument('--ann_file', dest='ann_file',
help='Annotations file',
default='../output/datasets/nucoco_sw_f/annotations/instances_train.json')
parser.add_argument('--imgs_dir', dest='imgs_dir',
help='Images directory',
default='../output/datasets/nucoco_sw_f/train')
args = parser.parse_args()
return args
if __name__ == '__main__':
args = parse_args()
coco = COCO_PLUS(args.ann_file, args.imgs_dir)
#fig, ax = plt.subplots(1, 1, figsize=(9, 9))
pprint(coco.cats)
input('something')
# Open samples from the NuScenes dataset
for i in tqdm(range(0, len(coco.dataset['images']))):
image_id = coco.dataset['images'][i]['id']
ann_ids = coco.getAnnIds(image_id)
anns = coco.loadAnns(ann_ids)
# print('Image ID: {}'.format(image_id))
# print('Ann IDs: {}'.format(ann_ids))
# print(anns[0])
img_path = os.path.join(args.imgs_dir, coco.imgs[image_id]["file_name"])