Vision-based 3D box container. The tvec is computed from projected center, depth, and intrinsics.

SE(3) rigid transform class that allows compounding of 6-DOF poses and provides common transformations that are commonly seen in geometric problem

A residual block that can optionally change the number of channels. :param channels: the number of input channels. :param emb_channels:

A sequential module that passes timestep embeddings to the children that support it as an extra input.

A Transform3d object encapsulates a batch of N 3D transformations, and knows how to transform points and normal vectors. Suppose that t is a

A downsampling layer with an optional convolution. :param channels: channels in the inputs and outputs. :param use_conv: a bool determin

Dummy class for backward-compatibility. Same as DetectionEval.

An upsampling layer with an optional convolution. :param channels: channels in the inputs and outputs. :param use_conv: a bool determini

Implementation that supports fp16 inputs but fp32 gains/biases.

ScanNet data. Generate scannet infos for scannet_converter. Args: root_path (str): Root path of the raw data. split (str): S

Adapted from detectron2: https://github.com/facebookresearch/detectron2/blob/master/detectron2/structures/image_list.py) Key differe

SUNRGBD data. Generate scannet infos for sunrgbd_converter. Args: root_path (str): Root path of the raw data. split (str): S

ScanNet dataset used to generate infos for semantic segmentation task. Args: data_root (str): Root path of the raw data. ann_file

Intersetion Over Union (IoU) loss which supports three different IoU computations: * IoU * Linear IoU * gIoU

A module which performs QKV attention. Matches legacy QKVAttention + input/ouput heads shaping

S3DIS data. Generate s3dis infos for s3dis_converter. Args: root_path (str): Root path of the raw data. split (str): Set spl

S3DIS dataset used to generate infos for semantic segmentation task. Args: data_root (str): Root path of the raw data. ann_file (

r"""Implements AdamW algorithm. Solve the bug of torch 1.8 The original Adam algorithm was proposed in `Adam: A Method for Stochastic Optimizatio

An attention block that allows spatial positions to attend to each other. Originally ported from here, but adapted to the N-d case. http

BBox3DL1Cost. Args: weight (int | float, optional): loss_weight

BEVFormer. Args: video_test_mode (bool): Decide whether to use temporal information during inference.

Attention with both self and cross Implements the decoder in DETR transformer. Args: return_intermediate (bool): Whether to retur

Head of Detr3D. Args: with_box_refine (bool): Whether to refine the reference points in the decoder. Defaults to False.

Implements decoder layer in DETR transformer. Args: attn_cfgs (list[`mmcv.ConfigDict`] | list[dict] | dict )): Configs for sel

BEVFormer. Args: video_test_mode (bool): Decide whether to use temporal information during inference.

The default version BEVFormer currently can not support FP16. We provide this version to resolve this issue.

Fixed Crop and then randim resize and flip the image. Note the flip requires to flip the feature in the network ida_aug_conf = { "reisz

Collect data from the loader relevant to the specific task. This is usually the last stage of the data loader pipeline. Typically keys is set

Default formatting bundle. It simplifies the pipeline of formatting common fields for voxels, including "proposals", "gt_bboxes", "gt_labels",

An attention module used in Deformable-Detr. `Deformable DETR: Deformable Transformers for End-to-End Object Detection. <https://arxiv.org/pd

r"""NuScenes Dataset. This datset only add camera intrinsics and extrinsics to the results.

r"""Monocular 3D detection on NuScenes Dataset. This class serves as the API for experiments on the NuScenes Dataset. Please refer to `NuScene

Implements the decoder in DETR3D transformer. Args: return_intermediate (bool): Whether to return intermediate outputs. coder_norm

Sampler that restricts data loading to a subset of the dataset. It is especially useful in conjunction with :class:`torch.nn.parallel.Distribu

# basic logic input_sequence = [a, b, c] # given a sequence of samples prev_bev = None for each in input_sequcene[:-1]

Random resize, Crop and flip the image Args: size (tuple, optional): Fixed padding size.

A wrapper for ``torch.nn.MultiheadAttention``. This module implements MultiheadAttention with identity connection, and positional encoding is

Computes one-to-one matching between predictions and ground truth. This class computes an assignment between the targets and the predictions b

A UNetModel that conditions on layout with an encoding transformer. The full UNet model with attention and timestep embedding. :param in

multi-modality fusion layer.

An attention module used in BEVFormer based on Deformable-Detr. `Deformable DETR: Deformable Transformers for End-to-End Object Detection. <ht

Base `TransformerLayer` for vision transformer. It can be built from `mmcv.ConfigDict` and support more flexible customization, for example, u

Bbox coder for NMS-free detector. Args: pc_range (list[float]): Range of point cloud. post_center_range (list[float]): Limit of th

Normalize the image. Added key is "img_norm_cfg". Args: mean (sequence): Mean values of 3 channels. std (sequence): Std values