def build_pred(self):
seq_data = []
if self.data_path == []:
print("Empty data")
return seq_data
for idx, file_path in enumerate(self.data_path):
frame_data = bh.load_json(file_path)
if idx % args.verbose_interval == 0 and idx != 0:
print('{} images.'.format(idx))
match_box = self.det_result[idx]
prediction_data = []
for i, box in enumerate(match_box):
prediction = bh.init_labels_format()
prediction['box2d']['x1'] = int(box[0])
prediction['box2d']['y1'] = int(box[1])
prediction['box2d']['x2'] = int(box[2])
prediction['box2d']['y2'] = int(box[3])
prediction['box2d']['confidence'] = float(box[6])
# 3D box projected center
prediction['box3d']['xc'] = int(box[4])
prediction['box3d']['yc'] = int(box[5])
del prediction['poly2d']
del prediction['attributes']['trafficLightColor']
del prediction['attributes']['areaType']
del prediction['attributes']['laneDirection']
del prediction['attributes']['laneStyle']
del prediction['attributes']['laneTypes']
prediction_data.append(prediction)
frame_data['prediction'] = prediction_data
print("Frame {}, GT: {} Boxes, PD: {} Boxes".format(
idx,
len(frame_data['labels']),
len(frame_data['prediction'])))
del frame_data['labels']
filename = os.path.join(
DATASET.PRED_PATH,
frame_data['videoName'],
os.path.basename(file_path))
if not os.path.exists(os.path.dirname(filename)):
os.mkdir(os.path.dirname(filename))
bh.dump_json(filename, frame_data)
seq_data.append(filename)
print("Saving {} frames...".format(idx))
return seq_data
def format_data2bdd(self):
dataset_list = []
for d_idx in range(len(self.data_idx)):
dataset_list.append(
self.get_data2bdd(self.data_idx[d_idx], self.data_label[d_idx],
self.box_idx[d_idx], self.det_result[d_idx]))
print("Saving {} with {} sequences...".format(save_name,
len(dataset_list)))
if not args.is_pred: bh.dump_json(save_name, dataset_list)
print("Saved")
# Load data
print("Load label file from path: {}".format(DATASET.PRED_PATH))
folders, data_path = load_label_path(DATASET.PRED_PATH)
for seq_idx, folder in enumerate(folders):
pkl_path = '{ROOT}{SESS}_{EP}_{DT}_{PH}_set/{SESS}_{EP}_{' \
'SQ}_bdd_roipool_output.pkl'.format(
**{'ROOT': args.root, 'SESS': args.session, 'EP': args.epoch,
'DT': args.set, 'PH': args.split, 'SQ': folder})
save_path = '{ROOT}{SESS}_{EP}_{DT}_{PH}_set/{SQ}_bdd_3d.json'.format(
**{'ROOT': args.root, 'SESS': args.session, 'EP': args.epoch,
'DT': args.set, 'PH': args.split, 'SQ': folder})
det_pred = load_single_frame_result(pkl_path)
hypos = []
for fr_idx, frame_path in enumerate(data_path[folder]):
det_placeholder = bh.load_json(frame_path)
hypo = convert_app(det_placeholder, det_pred[fr_idx])
hypos.append(hypo)
print("Saving updated tracking results with {} frames at {}...".format(
len(hypos), save_path))
bh.dump_json(save_path, hypos)
def dump_output_json(filename, frame_data):
if not os.path.exists(os.path.dirname(filename)):
print("Create {} ...".format(os.path.dirname(filename)))
os.makedirs(os.path.dirname(filename))
bh.dump_json(filename, frame_data)
def get_data2bdd(self, data_idx, data_label, box_idx, det_result):
seq_data = []
if data_idx == []:
print("Empty data")
return seq_data
for idx, im_path in enumerate(data_idx):
check_avail = box_idx[idx]
obj = data_label[im_path]
if idx % args.verbose_interval == 0 and idx != 0:
print('{} images.'.format(idx))
frame_data = bh.init_frame_format()
frame_data['name'] = im_path
del frame_data['url']
frame_data['videoName'] = obj['vid_name']
frame_data['resolution']['height'] = im_size_h
frame_data['resolution']['width'] = im_size_w
if 'weather' in obj and 'timeofday' in obj:
frame_data['attributes']['weather'] = obj['weather']
frame_data['attributes']['timeofday'] = obj['timeofday']
else:
del frame_data['attributes']
frame_data['frameIndex'] = idx
if 'cali' in obj:
frame_data['intrinsics']['cali'] = obj['cali']
frame_data['intrinsics']['focal'] = [
obj['cali'][0][0], obj['cali'][1][1]
]
frame_data['intrinsics']['center'] = [
obj['cali'][0][2], obj['cali'][1][2]
]
else:
frame_data['intrinsics']['cali'] = [[
focal, 0, im_size_w // 2, 0
], [0, focal, im_size_h // 2, 0], [0, 0, 1, 0]]
frame_data['intrinsics']['focal'] = [focal, focal]
frame_data['intrinsics']['center'] = [
im_size_w // 2, im_size_h // 2
]
frame_data['intrinsics']['fov'] = obj['fov']
frame_data['intrinsics']['nearClip'] = obj['nearClip']
# Camera relative rotation from previous frame (in rad)
# Camera relative location in world coordinate (in meter)
frame_data['extrinsics']['rotation'] = obj['pose']['rotation']
frame_data['extrinsics']['location'] = obj['pose']['position']
frame_data['timestamp'] = obj['timestamp']
if args.is_pred:
match_box = det_result[idx]
prediction_data = []
for i, box in enumerate(match_box):
prediction = bh.init_labels_format()
prediction['box2d']['x1'] = int(box[0])
prediction['box2d']['y1'] = int(box[1])
prediction['box2d']['x2'] = int(box[2])
prediction['box2d']['y2'] = int(box[3])
prediction['box2d']['confidence'] = float(box[6])
# 3D box projected center
prediction['box3d']['xc'] = int(box[4])
prediction['box3d']['yc'] = int(box[5])
del prediction['poly2d']
prediction_data.append(prediction)
frame_data['prediction'] = prediction_data
print("Frame {}, GT: {} Boxes, PD: {} Boxes".format(
idx, len(frame_data['labels']),
len(frame_data['prediction'])))
filename = os.path.join(DATASET.PRED_PATH,
os.path.splitext(im_path)[0] + '.json')
del frame_data['labels']
if not os.path.exists(os.path.dirname(filename)):
os.mkdir(os.path.dirname(filename))
bh.dump_json(filename, frame_data)
else:
frame_data['labels'] = self.get_data_from_id_bdd(
obj, check_avail)
del frame_data['prediction']
print("Frame {}, GT: {} Boxes".format(
idx, len(frame_data['labels'])))
filename = os.path.join(DATASET.LABEL_PATH,
os.path.splitext(im_path)[0] + '.json')
bh.dump_json(filename, frame_data)
seq_data.append(filename)
filename = os.path.join(DATASET.LABEL_PATH,
frame_data['videoName'] + '_bdd.json')
print("Saving {} with {} frames...".format(filename, idx))
if not args.is_pred: bh.dump_json(filename, seq_data)
return seq_data