Add shared or separable branch.

Affine transform bbox points to input image. Args: points (np.ndarray): Points to be transformed. shape: (N,

Draw bbox on visualizer and change the color of points inside bbox3d. Args: bbox3d (numpy.array | torch.tensor, shape=[M, 7]):

Get axis_align_matrix from info. If not exist, return identity mat. Args: info (dict): one data info term. Returns:

Get affine transform matrix. Args: center (tuple): Center of current image. scale (tuple): Scale of current image

Set flag according to image aspect ratio. Images with aspect ratio greater than 1 will be set as group 1, otherwise group 0. In 3D

Add bounding box to visualizer. Args: bbox3d (numpy.array, shape=[M, 7]): 3D bbox (x, y, z, x_size, y_size, z

3D NMS for aligned boxes. Args: boxes (torch.Tensor): Aligned box with shape [n, 6]. scores (torch.Tensor): Scores of each bo

Generate anchors in a single range. Args: feature_size (list[float] | tuple[float]): Feature map size. It is

Convert continuous angle to a discrete class and a residual. Convert continuous angle to a discrete class and a small regression n

Calculate nearest 3D IoU. Note: This function first finds the nearest 2D boxes in bird eye view (BEV), and then calculates th

Build RoI feature extractor.

Convert points in camera coordinate to lidar coordinate. Note: This function is for KITTI only. Args: points (np.ndarr

Inverse function to angle2class. Args: angle_cls (torch.Tensor): Angle class to decode. angle_res (torch.Tensor):

Classify each points.

Decode probabilistic depth map. Args: depth_cls_preds (torch.Tensor): Depth probabilistic map in shape [...,

Decode yaw angle and change it from local to global.i. Args: bbox (torch.Tensor): Bounding box predictions in shape

Encode points with backbone and decode into a semantic segmentation map of the same size as input. Args: points (torc

Enlarge the length, width and height boxes. Args: extra_width (float | torch.Tensor): Extra width to enlarge the box.

Kitti eval. support 2d/bev/3d/aos eval. support 0.5:0.05:0.95 coco AP. Args: gt_annos (dict): Must from get_label_annos() in kitti_com

Extract bounding boxes from 2d detector. Args: img (torch.Tensor): of shape (N, C, H, W) encoding input images.

Extract features from images and points.

Extract features of points.

During inference, the functionary of batch norm layers is turned off but only the mean and var alone channels are used, which exposes the chance

Get radius of gaussian. Args: det_size (tuple[torch.Tensor]): Size of the detection result. min_overlap (float, optional): Ga

Get anchors according to feature map sizes. Args: featmap_sizes (list[tuple]): Multi-level feature map sizes. inp

Get annotation info according to the given index. Args: index (int): Index of the annotation data to get. Returns:

Generate bboxes from vote head predictions. Args: points (torch.Tensor): Input points. bbox_preds (dict): Predict

Get data info according to the given index. Args: index (int): Index of the sample data to get. Returns:

Get data info according to the given index. Args: index (int): Index of the sample data to get. Returns:

Get data infos. This method gets information from the raw data. Args: num_workers (int, optional): Number of thread

KITTI annotation format version 2: { [optional]points: [N, 3+] point cloud [optional, for kitti]image: { ima

Only keep loading image, points and annotations related configuration. Args: pipeline (list[dict] | list[:obj:`Pipeline`]):

Get root logger and add a keyword filter to it. The logger will be initialized if it has not been initialized. By default a StreamHandler

Compute surface and line center of bounding boxes. Returns: torch.Tensor: Surface and line center of bounding boxes.

Inference point cloud with the detector. Args: model (nn.Module): The loaded detector. pcd (str): Point cloud files. R

Inverse function of sigmoid. Args: x (Tensor): The tensor to do the inverse. eps (float): EPS avoid numerical

Limit the yaw to a given period and offset. Args: offset (float, optional): The offset of the yaw. Defaults to 0.5.

Make auxiliary points for loss computation. Args: source_tensor (torch.Tensor): (M, C) features to be propigated.

Plot the boundary lines of 3D rectangular on 2D images. Args: img (numpy.array): The numpy array of image. num_rects (int): N

Initialization before data preparation. Args: results (dict): Dict before data preprocessing. - img_fields

r"""Randomly sample seeds from all imvotes. Modified from `<https://github.com/facebookresearch/imvotenet/blob/a8856345146bacf29a57266a2f0b8744

Setup multi-processing environment variables.

Test without augmentation, stage 2. Args: points (list[torch.Tensor], optional): Elements in the list should

RPN test function.

Call function to apply noise to each ground truth in the scene. Args: input_dict (dict): Result dict from loading pipeline.

Pillar Feature Net Layer. The Pillar Feature Net could be composed of a series of these layers, but the PointPillars paper results

int: Number of points in the current object.

Add shared or separable branch. Args: in_channels (int): Input feature channel. conv_channels (tuple): Middle fea

Compute assignment by checking whether point is inside bbox. Args: bboxes_3d (BaseInstance3DBoxes): Instance of bounding boxes

Compute assignment by checking whether point is inside bbox. Args: bboxes_3d (:obj:`BaseInstance3DBoxes`): Instance of boundin

Forward function of roi_extractor and bbox_head used in both training and testing. Args: seg_feats (torch.Tensor): Po

Forward function of roi_extractor and bbox_head used in both training and testing. Args: features (torch.Tensor): Bac

Transform predicted bbox parameters to bbox. Args: points (Tensor): Final locations of shape (N, 3). bbox_pred (T

Transform predicted bbox parameters to bbox. Args: points (Tensor): Final locations of shape (N, 3) bbox_pred (Te

Transform box to the axis-aligned or rotated iou loss format. Args: bbox (Tensor): 3D box of shape (N, 6) or (N, 7).

Retrieve object orientation. Args: ori_vector (Tensor): Local orientation in [sin, cos] format. shape: (N, 2)

Evaluation for a single model in nuScenes protocol. Args: result_path (str): Path of the result file. logger (log

Evaluation for a single model in nuScenes protocol. Args: result_path (str): Path of the result file. logger (log

Evaluation for a single model in Lyft protocol. Args: result_path (str): Path of the result file. logger (logging

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

Convert the results to the standard format. Args: results (list[dict]): Testing results of the dataset. jsonfile_

Convert the results to the standard format. Args: results (list[dict]): Testing results of the dataset. jsonfile_

Convert the results to the standard format. Args: results (list[dict]): Testing results of the dataset. jsonfile_

Return the channel number of classification outputs.

Return the channel number of classification outputs.

Compute the equation of the plane. Args: vector1 (torch.Tensor): Parallel vector of the plane. vector2 (torch.Ten

Generate final locations. Args: features (list[Tensor]): Feature tensors for all feature levels. Returns:

Get reference point to calculate affine transform matrix. While using opencv to calculate the affine matrix, we need at least thre

Return the channel number of regression outputs.

Return the channel number of regression outputs.

Make fully connect layers. Args: fc_cfg (dict): Config of fully connect. input_channels (int): Input channels for

Perform feature aggregation using pooling operation. Args: features (torch.Tensor): (B, C, N, K) Features of

Prepare input point pairs features for self.ScoreNet. Args: points_xyz (torch.Tensor): (B, 3, npoint, K) Coor

Removes point too close within a certain radius from origin. Args: points (np.ndarray | :obj:`BasePoints`): Sweep points.

Removes point too close within a certain radius from origin. Args: points (np.ndarray | :obj:`BasePoints`): Sweep points.

Rotate 2D boxes. Args: corners (np.ndarray): Corners of boxes. angle (float): Rotation angle. rot_mat_T (np.ndarray)

Perform point sampling based on inputs. If `indices` is specified, directly sample corresponding points. Else if `target_xyz` is s

Sample points for each points subset. Args: points (np.ndarray): Points subset to be sampled. sampler (VoxelGener

Select transform. Args: transform (np.ndarray): Transforms to select from. indices (np.ndarray): Mask to indicate which trans

Split coordinates and features of input points. Args: points (torch.Tensor): Point coordinates with features,

Upsample valid mask predictions. Args: valid_pred (Tensor): Valid mask prediction. features (Tensor): Feature ten

Export oriented (around Z axis) scene bbox to meshes. Args: scene_bbox(list[ndarray] or ndarray): xyz pos of center and 3

Add segmentation mask to visualizer via per-point colorization. Args: seg_mask_colors (numpy.array, shape=[N, 6]):