sam3-traffic-dataset
收藏资源简介:
sam3_traffic_dataset是一个采用YOLO格式的交通数据集,主要用于图像分割和物体检测任务。数据集包含机器生成的标注信息,分为训练集和验证集,分别存放在images和labels目录下。数据集涵盖六种交通相关对象类别:汽车、卡车、公交车、摩托车、自行车和行人。YOLO配置文件指定了训练和验证图像的路径,并设置了类别数量(nc: 6)。该数据集适用于使用Ultralytics框架进行模型训练和评估。
sam3_traffic_dataset is a traffic dataset in YOLO format, primarily intended for image segmentation and object detection tasks. The dataset includes machine-generated annotation data, and is split into training and validation sets, which are stored in the `images` and `labels` directories respectively. It covers six traffic-related object categories: car, truck, bus, motorcycle, bicycle, and pedestrian. The YOLO configuration file specifies the paths of training and validation images, and sets the number of categories (nc: 6). This dataset is suitable for model training and evaluation using the Ultralytics framework.
SAM3交通数据集概述
数据集基本信息
- 任务类别:图像分割、目标检测
- 标注创建方式:机器生成
- 许可证:其他
- 数据集名称:sam3_traffic_dataset
数据集描述
这是一个YOLO格式的交通数据集,源自特定项目。
数据集结构
dataset.yamlimages/trainimages/vallabels/trainlabels/val
YOLO配置
- 训练集路径:
images/train - 验证集路径:
images/val - 类别数量:
6
目标类别
- car
- truck
- bus
- motorcycle
- bicycle
- pedestrian
快速使用指南(Ultralytics)
bash
从Hugging Face下载仓库快照
python -c "from huggingface_hub import snapshot_download; snapshot_download(repo_id=iantc104/sam3-traffic-dataset, repo_type=dataset, local_dir=hf_dataset)"
使用本地快照进行训练
yolo task=segment mode=train model=yolo11n-seg.pt data=hf_dataset/dataset.yaml




