Sampling all categories of bboxes. Args: gt_bboxes (np.ndarray): Ground truth bounding boxes. gt_labels (np.ndarr

Scale the points with horizontal and vertical scaling factors. Args: scale_factors (float): Scale factors to scale the points.

Set random seed. Args: seed (int): Seed to be used. deterministic (bool): Whether to set the deterministic option for

Set template array. Args: array (tuple | list | int | float | np.ndarray | torch.Tensor): Template array.

Visualize the points cloud. Args: save_path (str, optional): path to save image. Default: None.

Transform the coordinates with matrix T. Args: T (np.ndarray): Transformation matrix. x(float): Coordinate in x a

Data preparation for voxel pooling. Args: coor (torch.tensor): Coordinate of points in the lidar space in sha

Apply hard voxelization to points.

Call function to load multiple types annotations. Args: results (dict): Result dict from :obj:`mmdet3d.CustomDataset`.

Call function to collect keys in results. The keys in ``meta_keys`` will be converted to :obj:`mmcv.DataContainer`. Args:

Affine transform bboxes to input image. Args: results (dict): Result dict from loading pipeline. matrix (np.ndarr

Assign and sample proposals for training. Args: proposal_list (list[dict]): Proposals produced by RPN. gt_bboxes_

Assign and sample proposals for training. Args: proposal_list (list[dict]): Proposals produced by RPN. gt_bboxes_

Run forward function and calculate loss for auxiliary head in training.

Forward training function of roi_extractor and bbox_head. Args: seg_feats (torch.Tensor): Point-wise semantic features.

Forward training function of roi_extractor and bbox_head. Args: features (torch.Tensor): Backbone features with depth and \

Build the default pipeline for this dataset.

Build the default pipeline for this dataset.

Build the default pipeline for this dataset.

Make ann_files as list/tuple.

Check if rotation matrix is valid for self.rotation_axis. Args: rot_mat (np.ndarray): Rotation matrix to be applied.

Make scene_idxs as list/tuple.

Convert class_id in loaded segmentation mask to label.

Convert class_id in loaded segmentation mask to label.

Transform depth offset to depth.

Transform dimension offsets to dimension according to its category. Args: labels (Tensor): Each points' category id.

Run forward function and calculate loss for decode head in inference.

Run forward function and calculate loss for decode head in training.

Retrieve objects location in camera coordinate based on projected points. Args: points (Tensor): Projected points on

Draw bbox on visualizer and change the color or points inside bbox3d with indices. Args: bbox3d (numpy.array | torch.tensor, shap

Load data using input pipeline and extract data according to key. Args: index (int): Index for accessing the target data.

Load data using input pipeline and extract data according to key. Args: index (int): Index for accessing the target data.

Extract inputs from features dictionary. Args: feat_dict (dict): Feature dict from backbone, which may contain

Extract inputs from features dictionary. Args: feat_dict (dict): Feature dict from backbone. Returns:

Extract inputs from features dictionary. Args: feat_dict (dict): Feature dict from backbone. Returns:

Extract inputs from features dictionary. Args: feat_dict (dict): Feature dict from backbone. Returns:

Generate the train/val infos from the raw data. Args: nusc (:obj:`NuScenes`): Dataset class in the nuScenes dataset. train_sc

Generate the train/val infos from the raw data. Args: lyft (:obj:`LyftDataset`): Dataset class in the Lyft dataset. train_sce

Filter images without ground truths.

Forward pass per level. Args: x (SparseTensor): Per level neck output tensor. scale (mmcv.cnn.Scale): Per level m

Gather feature map. Given feature map and index, return indexed feature map. Args: feat (torch.tensor): Feature map

Generate boxes for all scenes. Args: center_preds (list[list[Tensor]]): Centerness predictions for all scenes

Generate boxes for a single scene. Args: center_preds (list[Tensor]): Centerness predictions for all levels. bbox

Transform outputs for a single batch item into bbox predictions. Args: cls_scores (list[Tensor]): Box scores for a single scal

Transform outputs for a single batch item into bbox predictions. Args: cls_scores (list[Tensor]): Box scores for a single scal

Generate boxes for a single scene. Args: center_preds (list[Tensor]): Centerness predictions for all levels. bbox

Decode bboxes. Args: cls_scores (Tensor): Outputs from the classification head, \ shape [num_query, cls_out_cha

Compute point centerness w.r.t containing box. Args: face_distances (Tensor): Face distances of shape (B, N, 6),

Compute point centerness w.r.t containing box. Args: face_distances (Tensor): Face distances of shape (B, N, 6),

Return the channel number of classification outputs.

Calculate distances from point to box faces. Args: points (Tensor): Final locations of shape (N_points, N_boxes, 3).

Calculate distances from point to box faces. Args: points (Tensor): Final locations of shape (N_points, N_boxes, 3).

Get data loading pipeline in self.show/evaluate function. Args: pipeline (list[dict]): Input pipeline. If None is given,

Get data loading pipeline in self.show/evaluate function. Args: pipeline (list[dict]): Input pipeline. If None is given,

Get points of a single scale level.

Get point cloud filename according to the given index. Args: index (int): Index of the point cloud file to get. Ret

Return the channel number of regression outputs.

Load annotations from ann_file. Args: ann_file (str): Path of the annotation file. Returns: list[dict]

Compute targets for final locations for a single scene. Args: points (list[Tensor]): Final locations for all levels.

Compute targets for final locations for a single scene. Args: points (list[Tensor]): Final locations for all levels.

Initialize the target assigner and sampler of the head.

Initialize ``auxiliary_head``

Initialize classification conv layers of the head.

Initialize ``decode_head``

Initialize edge fusion module for feature extraction.

Initialize neural network layers of the head.

Initialize layers of the head.

Initialize neural network layers of the head.

Initialize layers. Args: in_channels (tuple[int]): Number of channels in input tensors. out_channels (int): Numbe

Initialize ``loss_regularization``

Initialize predictor layers of the head.

Initialize predictor layers of the head.

Initialize bbox regression conv layers of the head.

Generating model input. Generate input by subtracting patch center and adding additional features. Currently support colors an

Generating model input. Generate input by subtracting patch center and adding additional features. Currently support colors an

Private function to load label annotations. Args: results (dict): Result dict from :obj:`mmdet3d.CustomDataset`. Re

Private function to load 3D bounding box annotations. Args: results (dict): Result dict from :obj:`mmdet3d.CustomDataset`.

Private function to load 2.5D bounding box annotations. Args: results (dict): Result dict from :obj:`mmdet3d.CustomDataset`.

Private function to load label annotations. Args: results (dict): Result dict from :obj:`mmdet3d.CustomDataset`. Re

Private function to load 3D mask annotations. Args: results (dict): Result dict from :obj:`mmdet3d.CustomDataset`.