example = { 'x.lidar.fp': _paired_frame[0], 'y.camera.fp': _paired_frame[1], 'x.camera.fps': [x[1] for x

example = { 'x.lidar.fp': _paired_frame[0], 'y.camera.fp': _paired_frame[1], 'x.camera.fps': [x[1] for x

example = { 'x.lidar.fp': _paired_frame[0], 'y.camera.fp': _paired_frame[1], 'x.camera.fps': [x[1] for x

example = { 'x.lidar.fp': _paired_frame[0], 'y.camera.fp': _paired_frame[1], 'x.camera.fps': [x[1] for x

example = { 'x.lidar.fp': _paired_frame[0], 'y.camera.fp': _paired_frame[1], 'x.camera.fps': [x[1] for x

This is a helper function that will scan KITTI raw data folder and output stats :param data_fp: absolute folder path :return: - s

https://www.tensorflow.org/api_docs/python/tf/contrib/framework/get_variables_to_restore :param graph: the graph to be restored :param mo

:param config: :return: labelled_testing_data_dict = { "14330_5": { 'gt': 1,

:param config: :param data_info_synced: data_info = { lidar_sub_folder: {}, camera_sub_folder

:param config: :param data_info_synced: data_info = { lidar_sub_folder: {}, camera_sub_folder

:param config: :param data_info_synced: data_info = { lidar_sub_folder: {}, camera_sub_folde

:param config: :param data_info_synced: data_info = { lidar_sub_folder: {}, camera_sub_folde

:param config: :param data_info_synced: data_info = { lidar_sub_folder: {}, camera_sub_folde

Adding your customized init here :return:

Below we only import image_02, left rectified color image This is a helper function that will load all KITTI raw data into memory at once.

Below we only import image_02, left rectified color image This is a helper function that will load all KITTI raw data into memory at once.

:param ts: a sorted array of ts, such as 1583836591.1523867 :param offsets: csv file %time field.header.seq fiel

:param ts: a sorted array of ts, such as 1583836591.1523867 :param offsets: csv file %time field.header.seq fiel

:param ts: a sorted array of ts, such as 1583836591.1523867 :param offsets: csv file %time field.header.seq fiel

:param ts: a sorted array of ts, such as 1583836591.1523867 :param offsets: csv file %time field.header.seq fiel

:param ts: a sorted array of ts, such as 1583836591.1523867 :param offsets: csv file %time field.header.seq fiel

If you don't need to do any pre-process, return feed_dict :param feed_dict: :param tensor_dict: :return: feed_dict

Converge X from shape (B, H, W, num_frames x 4) to (B x num_frames, H, W, 4) so that a network can learn to inference on single (?, H

Converge X from shape (B, H, W, num_frames x 4) to (B x num_frames, H, W, 4) so that a network can learn to inference on single (?, H

Converge X from shape (B, H, W, num_frames x 4) to (B x num_frames, H, W, 4) so that a network can learn to inference on single (?, H

Converge X from shape (B, H, W, num_frames x 4) to (B x num_frames, H, W, 4) so that a network can learn to inference on single (?, H

Converge X from shape (B, H, W, num_frames x 4) to (B x num_frames, H, W, 4) so that a network can learn to inference on single (?, H

This function is a very generic function that - init graph - retrieve tfrecords init/batch get next ops -

Parse single example :param record: Single record in tfrecord :param feature_set: Dictionary of Features :return:

By default, we set the output name as $model_name/$_output_name in config. - We force you to rename your outputs as graceful name

this function read a rosbag and create a synthetic rosbag with mis-synchronization :param bag_fp: original bagpath :param max_shift: max

Freeze graph into single protobuf file. :param model_folder: Where the model is stored :param meta_info: Meta i

data_info = { "paired_fp_seq": { "2011_09_26_drive_0001_sync": [ [".../velody