Batch Norm Contrastive Head for YOLO-World using batch norm instead of l2-normalization Args: embed_dims (int): embed dim of text and

Contrastive Head for YOLO-World compute the region-text scores according to the similarity between image and text features Args: e

Max Sigmoid attention block.

Batch Norm Contrastive Head for YOLO-World using batch norm instead of l2-normalization Args: embed_dims (int): embed dim of text and

Max Sigmoid attention block.

Max Sigmoid attention block.

Sigmoid attention block.

A Base Transform of Multimodal multiple images mixed. Suitable for training on multiple images mixed data augmentation like mosaic and mixup.

Sigmoid-attention based CSP layer with two convolution layers.

Sigmoid-attention based CSP layer with two convolution layers.

Multi-modal dataset.

Multi-modal Mixed dataset. mix "detection dataset" and "caption dataset" Args: dataset_type (str): dataset type, 'detection' or 'capti

Mosaic augmentation. Given 4 images, mosaic transform combines them into one output image. The output image is composed of the parts from eac

Mosaic9 augmentation. Given 9 images, mosaic transform combines them into one output image. The output image is composed of the parts from ea

Pseudo Language Backbone Args: text_embed_path (str): path to the text embedding file

Sigmoid-attention based CSP layer with two convolution layers.

Sigmoid-attention based CSP layer with two convolution layers.

Implementation of YOLO World Series

TensorRT NMS operation.

Objects365 v1 dataset for detection.

Rewrite collate_fn to get faster training speed. Note: It must be used together with `mmyolo.datasets.utils.yolow_collate`

Implementation of YOLOW Series

Path Aggregation Network used in YOLO World v8.

YOLO-World Head

Head Module for YOLO-World Args: embed_dims (int): embed dim for text feautures and image features use_bn_head (bool): use batch

Implementation of YOLO World Series

Path Aggregation Network used in YOLO World Following YOLOv8 PAFPN, including text to image fusion

YOLO World v5 constructor for optimizers.

MixUp data augmentation for YOLOX. .. code:: text mixup transform +---------------+--------------+

Mixed grounding dataset.

Dataset for YOLOv5 LVIS Dataset. We only add `BatchShapePolicy` function compared with Objects365V1Dataset. See `mmyolo/datasets/utils.py#Bat

Mixed grounding dataset.

MixUp data augmentation for YOLOv5. .. code:: text The mixup transform steps are as follows: 1. Another random image is picked by d

Dataset for YOLOv5 VOC Dataset. We only add `BatchShapePolicy` function compared with Objects365V1Dataset. See `mmyolo/datasets/utils.py#Batc

Dataset for YOLOv5 VOC Dataset. We only add `BatchShapePolicy` function compared with Objects365V1Dataset. See `mmyolo/datasets/utils.py#Batc

Dataset for YOLOv5 VOC Dataset. We only add `BatchShapePolicy` function compared with Objects365V1Dataset. See `mmyolo/datasets/utils.py#Batc