Add shared or separable branch. convs -> avg pool (optional) -> fcs

Box head forward function used in both training and testing. Returns all regression, classification results and a intermediate feature.

Decode init_proposal_bboxes according to the size of images and expand dimension of init_proposal_features to batch_size. Args:

Create a superset of inputs needed to run test or train batches. Args: input_shape (tuple): input batch dimensions n

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

Gather feature according to index. Args: feat (Tensor): Target feature map. ind (Tensor): Target coord index.

Grab configs necessary to create a pts_bbox_head. These are deep copied to allow for safe modification of parameters without influencing othe

Initialize classification conv layers of the head.

Initialize bbox regression conv layers of the head.

Mask head forward function used in both training and testing.

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

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

Add ground truth as assigned results. Args: gt_labels (torch.Tensor): Labels of gt boxes

Convert the rotation difference to difference in sine function. Args: boxes1 (torch.Tensor): Original Boxes in shape (NxC), where

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

Get anchor centers from anchors. Args: anchors (Tensor): Anchor list with shape (N, 4), "xyxy" format. Returns:

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

Assign w.r.t. the overlaps of bboxes with gts. Args: overlaps (Tensor): Overlaps between k gt_bboxes and n bboxes,

Async inference image(s) with the detector. Args: model (nn.Module): The loaded detector. img (str | ndarray): Either image files

Test det bboxes with test time augmentation. Args: feats (list[Tensor]): the outer list indicates test-time augme

Decode bounding box based on distances. Args: points (Tensor): Shape (n, 2), [x, y]. bbox (Tensor): Shape (n, 4), "xyxy" format

The key correspondence from bboxes to labels, masks and segmentations.

Convert 2D detection results to kitti format for evaluation and test submission. Args: net_outputs (list[np.ndarray]): Li

Convert a list of bounding boxes to roi format. Args: bbox_list (list[torch.Tensor]): A list of bounding boxes corresponding

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

Convert bbox coordinates from (x1, y1, x2, y2) to (cx, cy, w, h). Args: bbox (Tensor): Shape (n, 4) for bboxes. Returns: Ten

Build neck.

Classification-Aware Regression Loss (CARL). Args: cls_score (Tensor): Predicted classification scores. labels (Tensor): Targets

Convert kitti locations, dimensions and angles to corners. format: center(xy), dims(xy), angles(clockwise when positive) Args: center

Check if norm layer is in correct train state.

Circular NMS. An object is only counted as positive if no other center with a higher confidence exists within a radius r using a bird-eye

Async context manager that waits for work to complete on given CUDA streams.

Convert the torch forward outputs containing tuple or list to a list only containing torch.Tensor. Args: output (list(Tensor) | tuple

Generate relative box corners based on length per dim and origin point. Args: dims (np.ndarray, shape=[N, ndim]): Array of length per dim

Given the raw data, generate the ground truth database. Args: dataset_class_name （str): Name of the input dataset. data_path (str

see :func:`BaseInstanceMasks.crop`

See :func:`BaseInstanceMasks.crop_and_resize`.

Get current learning rates. Returns: list[float] | dict[str, list[float]]: Current learning rates of all param gr

Map from level to values.

Calculate recalls. Args: gts (list[ndarray]): a list of arrays of shape (n, 4) proposals (list[ndarray]): a list of arrays of sha

Directly extract features from the backbone+neck.

Extract point and image features of multiple samples.

Flip the boxes in BEV along given BEV direction. In LIDAR coordinates, it flips the y (horizontal) or x (vertical) axis. Args:

Format the results to json (standard format for COCO evaluation). Args: results (list[dict]): Testing results of the dataset.

Format the results to json (standard format for COCO evaluation). Args: results (list[tuple | numpy.ndarray]): Testing results of

Args: img (Tensor): Input images of shape (N, C, H, W). Typically these should be mean centered and std scaled.

Generate voxels given points.

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

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

Get category ids of repeat dataset by index. Args: idx (int): Index of data. Returns: list[int]: All categor

Get COCO category ids by index. Args: idx (int): Index of data. Returns: list[int]: All categories in the im

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

Generate targets of vote head. Args: points (list[torch.Tensor]): Points of each batch. gt_bboxes_3d (list[:obj:`Base

Generate targets of proposal module. Args: points (list[torch.Tensor]): Points of each batch. gt_bboxes_3d (list[:obj

A wrapper for computing ATSS and FCOS targets for points in multiple images. Args: cls_scores (list[Tensor]): Box iou-awa

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

Indoor Evaluation. Evaluate the result of the detection. Args: gt_annos (list[dict]): Ground truth annotations. dt_annos (li

Initialize a detector from config file. Args: config (str or :obj:`mmcv.Config`): Config file path or the config object.

Initialize model weights.

Initialize the weights in detector. Args: pretrained (str, optional): Path to pre-trained weights. Defaults to No

Importance-based Sample Reweighting (ISR_P), positive part. Args: cls_score (Tensor): Predicted classification scores. bbox_pred

Map from level to values based on max_value.

Multi-class nms in single batch. Args: obj_scores (torch.Tensor): Objectness score of bounding boxes. sem_scores (tor

Find boxes that are non-empty. A box is considered empty, if either of its side is no larger than threshold. Args:

nn.Module: normalization layer after the first convolution layer

Prepare data related to nuScenes dataset. Related data consists of '.pkl' files recording basic infos, 2D annotations and groundtruth databas

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

Find points that are in boxes (CUDA) Args: points (torch.Tensor): [B, M, 3], [x, y, z] in LiDAR coordinate boxes (torch.Tensor):

Find points that are in boxes (CUDA) Args: points (torch.Tensor): [B, M, 3], [x, y, z] in LiDAR coordinate boxes (torch.Tensor):

Get training data and annotations after pipeline. Args: idx (int): Index of data. Returns: dict: Training da

Randomly negate value based on random_negative_prob.

RetinanNet Head Config.

Sample positive and negative bboxes. This is a simple implementation of bbox sampling given candidates, assigning results and ground

Sample features from single level image feature map. Args: img_feats (torch.Tensor): Image feature map in shape (

Save checkpoint to file. The checkpoint will have 4 fields: ``meta``, ``state_dict`` and ``optimizer``, ``amp``. By default ``meta`` will con

Expand an array of boxes by a given scale. Args: bboxes (Tensor): Shape (m, 4) scale (float): The scale factor of bboxes Ret

Visualize the detection results on the image. Args: model (nn.Module): The loaded detector. img (str or np.ndarray): Image filena

Test only det bboxes without augmentation. Args: x (tuple[Tensor]): Feature maps of all scale level. img_metas (list[

Simple test for mask head without augmentation.