DATASET_DIR = args.dataset_root_dir
if args.dataset_split_file == '':
DATASET_SPLIT_FILE = os.path.join(
DATASET_DIR, '3DOP_splits/' + train_config['train_dataset'])
else:
DATASET_SPLIT_FILE = args.dataset_split_file
config_complete = load_config(args.config_path)
if 'train' in config_complete:
config = config_complete['train']
else:
config = config_complete
# input function ==============================================================
dataset = KittiDataset(os.path.join(DATASET_DIR, 'image/training/image_2'),
os.path.join(DATASET_DIR,
'velodyne/training/velodyne/'),
os.path.join(DATASET_DIR, 'calib/training/calib/'),
os.path.join(DATASET_DIR, 'labels/training/label_2'),
DATASET_SPLIT_FILE,
num_classes=config['num_classes'])
NUM_CLASSES = dataset.num_classes
if 'NUM_TEST_SAMPLE' not in train_config:
NUM_TEST_SAMPLE = dataset.num_files
else:
if train_config['NUM_TEST_SAMPLE'] < 0:
NUM_TEST_SAMPLE = dataset.num_files
else:
NUM_TEST_SAMPLE = train_config['NUM_TEST_SAMPLE']
BOX_ENCODING_LEN = get_encoding_len(config['box_encoding_method'])
box_encoding_fn = get_box_encoding_fn(config['box_encoding_method'])
else:
DATASET_SPLIT_FILE = args.dataset_split_file
if args.output_dir == '':
OUTPUT_DIR = os.path.join(args.checkpoint_path, './eval/')
else:
OUTPUT_DIR = args.output_dir
CHECKPOINT_PATH = args.checkpoint_path
CONFIG_PATH = os.path.join(CHECKPOINT_PATH, 'config')
assert os.path.isfile(CONFIG_PATH), 'No config file found in %s'
config = load_config(CONFIG_PATH)
# setup dataset ===============================================================
if IS_TEST:
dataset = KittiDataset(os.path.join(DATASET_DIR, 'image/testing/image_2'),
os.path.join(DATASET_DIR,
'velodyne/testing/velodyne/'),
os.path.join(DATASET_DIR, 'calib/testing/calib/'),
'',
num_classes=config['num_classes'],
is_training=False)
else:
dataset = KittiDataset(os.path.join(DATASET_DIR, 'image/training/image_2'),
os.path.join(DATASET_DIR,
'velodyne/training/velodyne/'),
os.path.join(DATASET_DIR, 'calib/training/calib/'),
os.path.join(DATASET_DIR,
'labels/training/label_2'),
DATASET_SPLIT_FILE,
num_classes=config['num_classes'])
NUM_TEST_SAMPLE = dataset.num_files
NUM_CLASSES = dataset.num_classes
args = parser.parse_args()
train_config = load_train_config(args.train_config_path)
DATASET_DIR = args.dataset_root_dir
if args.dataset_split_file == '':
DATASET_SPLIT_FILE = os.path.join(
DATASET_DIR, './3DOP_splits/' + train_config['train_dataset'])
else:
DATASET_SPLIT_FILE = args.dataset_split_file
config_complete = load_config(args.config_path)
if 'train' in config_complete:
config = config_complete['train']
else:
config = config_complete
# input function ==============================================================
dataset = KittiDataset(os.path.join(DATASET_DIR,
'velodyne/training/velodyne/'),
os.path.join(DATASET_DIR, 'labels/training/label_2'),
num_classes=config['num_classes'])
NUM_CLASSES = dataset.num_classes
if 'NUM_TEST_SAMPLE' not in train_config:
NUM_TEST_SAMPLE = dataset.num_files
else:
if train_config['NUM_TEST_SAMPLE'] < 0:
NUM_TEST_SAMPLE = dataset.num_files
else:
NUM_TEST_SAMPLE = train_config['NUM_TEST_SAMPLE']
BOX_ENCODING_LEN = get_encoding_len(config['box_encoding_method'])
box_encoding_fn = get_box_encoding_fn(config['box_encoding_method'])
box_decoding_fn = get_box_decoding_fn(config['box_encoding_method'])
required=True)
flags = parser.parse_args()
hyper_params = json.loads(open("semantic_no_color.json").read())
# Create output dir
sparse_output_dir = os.path.join("result", "sparse")
dense_output_dir = os.path.join("result", "dense")
os.makedirs(sparse_output_dir, exist_ok=True)
os.makedirs(dense_output_dir, exist_ok=True)
# Dataset
dataset = KittiDataset(
num_points_per_sample=hyper_params["num_point"],
base_dir=flags.kitti_root,
dates=["2011_09_26"],
# drives=["0095", "0001"],
drives=["0095"],
box_size_x=hyper_params["box_size_x"],
box_size_y=hyper_params["box_size_y"],
)
# Model
max_batch_size = 128 # The more the better, limited by memory size
predictor = PredictInterpolator(
checkpoint_path=flags.ckpt,
num_classes=dataset.num_classes,
hyper_params=hyper_params,
)
# Init visualizer
dense_pcd = open3d.PointCloud()
default='../dataset/kitti/',
help='Path to KITTI dataset. Default="../dataset/kitti/"')
args = parser.parse_args()
eval_config = load_train_config(args.eval_config_path)
DATASET_DIR = args.dataset_root_dir
config_path = os.path.join(eval_config['train_dir'],
eval_config['config_path'])
while not os.path.isfile(config_path):
print('No config file found in %s, waiting' % config_path)
time.sleep(eval_config['eval_every_second'])
config = load_config(config_path)
if 'eval' in config:
config = config['eval']
dataset = KittiDataset(os.path.join(DATASET_DIR, 'velodyne/val/'),
os.path.join(DATASET_DIR, 'labels/val/'),
num_classes=config['num_classes'])
NUM_CLASSES = dataset.num_classes
print(dataset)
if 'NUM_TEST_SAMPLE' not in eval_config:
NUM_TEST_SAMPLE = dataset.num_files
else:
if eval_config['NUM_TEST_SAMPLE'] < 0:
NUM_TEST_SAMPLE = dataset.num_files
else:
NUM_TEST_SAMPLE = eval_config['NUM_TEST_SAMPLE']
BOX_ENCODING_LEN = get_encoding_len(config['box_encoding_method'])
box_encoding_fn = get_box_encoding_fn(config['box_encoding_method'])
box_decoding_fn = get_box_decoding_fn(config['box_encoding_method'])
def __len__(self):
return self.num_data
if __name__ == '__main__':
args = parse_args()
if torch.cuda.is_available() and not args.cuda:
print("WARNING: You might want to run with --cuda")
if not os.path.exists(args.output_dir):
os.makedirs(args.output_dir)
# dataset
if args.dataset == 'kitti':
dataset = KittiDataset(train=True)
train_size = len(dataset)
batch_sampler = sampler(train_size, args.bs)
dataloader = torch.utils.data.DataLoader(dataset, batch_size=args.bs,
sampler=batch_sampler, num_workers=args.num_workers)
elif args.dataset == 'scannet':
pass
# tensor placeholders
img = torch.FloatTensor(1)
z = torch.FloatTensor(1)
# network initialization
print('Initializing model...')
parser.add_argument("--save", action="store_true", default=False)
flags = parser.parse_args()
hyper_params = json.loads(open("semantic_no_color.json").read())
# Create output dir
sparse_output_dir = os.path.join("result", "sparse")
dense_output_dir = os.path.join("result", "dense")
os.makedirs(sparse_output_dir, exist_ok=True)
os.makedirs(dense_output_dir, exist_ok=True)
# Dataset
dataset = KittiDataset(
num_points_per_sample=hyper_params["num_point"],
base_dir="/home/ylao/data/kitti",
dates=["2011_09_26"],
# drives=["0095", "0001"],
drives=["0095"],
box_size_x=hyper_params["box_size_x"],
box_size_y=hyper_params["box_size_y"],
)
# Model
max_batch_size = 128 # The more the better, limited by memory size
predictor = PredictInterpolator(
checkpoint_path=flags.ckpt,
num_classes=dataset.num_classes,
hyper_params=hyper_params,
)
# Init visualizer
dense_pcd = open3d.PointCloud()
'Default="CHECKPOINT_PATH/eval/"')
args = parser.parse_args()
DATASET_DIR = args.dataset_root_dir
if args.output_dir == '':
OUTPUT_DIR = os.path.join(args.checkpoint_path, './eval/')
else:
OUTPUT_DIR = args.output_dir
CHECKPOINT_PATH = args.checkpoint_path
CONFIG_PATH = os.path.join(CHECKPOINT_PATH, 'config')
assert os.path.isfile(CONFIG_PATH), 'No config file found in %s'
config = load_config(CONFIG_PATH)
# setup dataset ===============================================================
dataset = KittiDataset(os.path.join(DATASET_DIR, 'velodyne/testing/'),
"",
num_classes=config['num_classes'],
is_training=False)
NUM_TEST_SAMPLE = dataset.num_files
NUM_CLASSES = dataset.num_classes
try:
for frame_idx in tqdm(range(0, NUM_TEST_SAMPLE)):
start_time = time.time()
# provide input ======================================================
cam_points = dataset.get_cam_points_in_image(
frame_idx, config['downsample_by_voxel_size'])
except:
print("ss")
import numpy as np
import open3d
import os
from dataset.kitti_dataset import KittiDataset
from sklearn.cluster import KMeans
from tqdm import tqdm
dataset = KittiDataset('../dataset/kitti/image/training/image_2',
'../dataset/kitti/velodyne/training/velodyne/',
'../dataset/kitti/calib/training/calib/',
'',
'../dataset/kitti/3DOP_splits/val.txt',
is_training=False)
downsample_rate = 2
output_dir = '../dataset/kitti/velodyne/training_downsampled_%d/velodyne/' % downsample_rate
for frame_idx in tqdm(range(0, dataset.num_files)):
velo_points = dataset.get_velo_points(frame_idx)
filename = dataset.get_filename(frame_idx)
xyz = velo_points.xyz
xyz_norm = np.sqrt(np.sum(xyz * xyz, axis=1, keepdims=True))
z_axis = np.array([[0], [0], [1]])
cos = xyz.dot(z_axis) / xyz_norm
kmeans = KMeans(n_clusters=64, n_jobs=-1).fit(cos)
centers = np.sort(np.squeeze(kmeans.cluster_centers_))
centers = [
-1,
] + centers.tolist() + [
1,
]
cos = np.squeeze(cos)