• 大数据-各类图像数据集下载地址


    各类图像数据集下载地址

    反代加速请参见另一篇

    COCO

    Images
    1. 官网
    2. https://cocodataset.org/#home
    3. 2014 Train images [83K/13GB]:
    4. http://images.cocodataset.org/zips/train2014.zip
    5. 2014 Val images [41K/6GB]:
    6. http://images.cocodataset.org/zips/val2014.zip
    7. 2014 Test images [41K/6GB]:
    8. http://images.cocodataset.org/zips/test2014.zip
    9. 2015 Test images [81K/12GB]:
    10. http://images.cocodataset.org/zips/test2015.zip
    11. 2017 Train images [118K/18GB]:
    12. http://images.cocodataset.org/zips/train2017.zip
    13. 2017 Val images [5K/1GB]:
    14. http://images.cocodataset.org/zips/val2017.zip
    15. 2017 Test images [41K/6GB]:
    16. http://images.cocodataset.org/zips/test2017.zip
    17. 2017 Unlabeled images [123K/19GB]:
    18. http://images.cocodataset.org/zips/unlabeled2017.zip

    Annotations

    1. 2014 Train/Val annotations [241MB]:
    2. http://images.cocodataset.org/annotations/annotations_trainval2014.zip
    3. 2014 Testing Image info [1MB]:
    4. http://images.cocodataset.org/annotations/image_info_test2014.zip
    5. 2015 Testing Image info [2MB]:
    6. http://images.cocodataset.org/annotations/image_info_test2015.zip
    7. 2017 Train/Val annotations [241MB]:
    8. http://images.cocodataset.org/annotations/annotations_trainval2017.zip
    9. 2017 Stuff Train/Val annotations [1.1GB]:
    10. http://images.cocodataset.org/annotations/stuff_annotations_trainval2017.zip
    11. 2017 Panoptic Train/Val annotations [821MB]:
    12. http://images.cocodataset.org/annotations/panoptic_annotations_trainval2017.zip
    13. 2017 Testing Image info [1MB]:
    14. http://images.cocodataset.org/annotations/image_info_test2017.zip
    15. 2017 Unlabeled Image info [4MB]:
    16. http://images.cocodataset.org/annotations/image_info_unlabeled2017.zip

    KITTI

    官网

    1. http://www.cvlibs.net/datasets/kitti/
    2. left color images of object data set (12 GB):
    3. https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_image_2.zip
    4. right color images, if you want to use stereo information (12 GB):
    5. https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_image_3.zip
    6. Velodyne point clouds, if you want to use laser information (29 GB):
    7. https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_velodyne.zip
    8. training labels of object data set (5 MB):
    9. https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_label_2.zip

    MPII

    官网

    1. http://human-pose.mpi-inf.mpg.de/#download
    2. Images (12.9 GB)
    3. https://datasets.d2.mpi-inf.mpg.de/andriluka14cvpr/mpii_human_pose_v1.tar.gz
    4. Annotations (12.5 MB)
    5. https://datasets.d2.mpi-inf.mpg.de/andriluka14cvpr/mpii_human_pose_v1_u12_2.zip

  • 相关阅读:
    MySQL数据库——存储过程-游标(介绍-声明游标、打开游标、获取游标记录、关闭游标,案例)
    【Linux】进程状态详解
    GFS 分布式文件系统
    C语言——内存函数
    Android C++系列:Linux文件IO操作(一)
    计算机系统概论
    最佳生物信息工作环境(2023年11月更新版)
    HTML期末学生大作业-拯救宠物网页作业html+css
    华为机试 - 处理器问题
    Android OpenGL ES 3.0 FBO 离屏渲染
  • 原文地址:https://blog.csdn.net/xiaoshun007/article/details/133069587