Generate bboxes from bbox head predictions. Args: preds_dicts (tuple[list[dict]]): Prediction results. img_metas (list

Get category distribution of single scene. Args: idx (int): Index of the data_info. Returns: dict[list]: for

Initialize weights of the DeformDETR head.

Initialize weights of the DeformDETR head.

Initialize the transformer weights.

Initialize weights of the depth head.

Initialize weights of the depth head.

Initialize weights of the depth head.

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

Loss function. Args: gt_bboxes_list (list[Tensor]): Ground truth bboxes for each image with shape (nu

Convert the model into training mode while keep normalization layer freezed.

bool: Whether the detector has a depth head.

bool: Whether the detector has a depth head.

bool: Whether the detector has a depth head.

bool: Whether the detector has a depth head.