NOTE: THIS IS A NUMPY FUNCTION input: dets, numpy array, shape in [batch x max_dets x dim] input: img_info, dict, necessary information o

:param data_folder: folder containing ground truth :param test_dataset_name: kitti_split1, kitti_split2 :param results_folder: folder wh

Hermite polynomial of order n calculated at X Args: n: int >= 0 X: np.array Output: Y: array of shape X.shape

Computes the amount of intersect between two different sets of boxes. Args: box_a (nparray): Mx4 boxes, defined by [x1, y1, x2, y2]

References: MonoPair: Monocular 3D Object Detection Using Pairwise Spatial Relationships, CVPR'20 Geometry and Uncertainty in Dee

Recursively make a directory structure even if missing. if delete_if_exist=True then we will delete it first which can be useful when be

Serialize an object to a provided file_path

Throws the outliers based on sorting frac = 0.75 suggests keep the first three quarters of the data and throw away the remain

dest_input is assumed to be copied of src_input

Inverse function to angle2class.