Download

You will receive the link after filling this form (https://forms.gle/P29nRMiWTGE9VXsT7). By downloading the dataset you accept the following License.

Dataset breakdown

  • 15 sequences of 12 human subjects in
  • 10 scenes in urban environments (1k – 30k $m^2$)
  • 100k+ frames multi-source data (20 Hz)
  • including 2D / 3D annotations and 3D scenes; 7 km+ human motions.

Every human subject signed permission to release their motion data for research purposes.

2D / 3D annotations and point clouds.

Human motions in reconstructed 3D scenes

Sequence name Traj. length ($m$) Area size ($m^2$) Frames Motions
001_campus_001 908 13,400 16,202 Jogging downhill, tying shoelaces, jumping
002_football_002 221 200 4,665 Juggling, passing, and shooting a football
003_street_002 291 1,600 6,496 Taking photos, putting on/taking off a bag
004_library_001 440 2,300 9,949 Borrowing books, reading, descending stairs
005_library_002 474 2,300 8,901 Looking up, reading, returning a book
006_library_003 477 2,300 8,386 Admiring paintings, throwing rubbish, greeting
007_garden_001 217 3,000 5,994 Raising hand, sitting on bench, going upstairs
008_running_001 392 8,500 2,000 Running
009_running_002 985 30,000 8,113 Running
010_park_001 642 9,300 12,445 Visiting a park, walking up a small hill
011_park_002 1,025 11,000 1,000 Buying drinks, trotting, drinking
012_square_001 264 3,200 6,792 Making phone calls, waving, drinking
013_sunlightRock001 386 1,900 10,116 Climbing stairs, taking photos, walking
014_garden_002 209 4,200 5,604 Stooping, crossing a bridge, sitting cross-legged
015_plaza_001 365 2,700 7,989 Admiring sculptures, eating

Data structure

# the data structure for every sequence
├── root_folder
   ├── lidar_data/
   |  ├── lidar_frames_rot/        
   |  |   └── '*.pcd'              # undistorted n frames point clouds in global coordinates
   |  ├── 'lidar_trajectory.txt'   # everyline: framenum X Y Z qx qy qz qw timestamp
   |  └── 'tracking_traj.txt'      # everyline: X Y Z framenum timestamp
   ├── mocap_data/
   |  └── '*_second.bvh'           # mocap data
   ├── rgb_data/
   |  └── '*.mp4'
   ├── '*_labels.pkl'              # all 2D/3D labels and origin human data
   └── 'dataset_params.json'       # meta info

Citation

@InProceedings{Dai_2023_CVPR,
    author    = {Dai, Yudi and Lin, Yitai and Lin, Xiping and Wen, Chenglu and Xu, Lan and Yi, Hongwei and Shen, Siqi and Ma, Yuexin and Wang, Cheng},
    title     = {SLOPER4D: A Scene-Aware Dataset for Global 4D Human Pose Estimation in Urban Environments},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2023},
    pages     = {682-692}
}

Commercial licensing

Please email us:
clwen@xmu.edu.cn
cwang@xmu.edu.cn