3D Human Pose Point Cloud Data of Light Detection and Ranging (LiDAR)
Description
This dataset contains 3D human pose point cloud data of four basic human poses demonstrated by one male talent. These basic human poses include hands-to-the-side, sit-down, squat-down, and stand-up. The total dataset consists of 1,400 3D human point cloud data, where 1,120 are training data and 280 are testing data. Each class consists of 280 data. The distribution of the training data and testing data is as follows: the training data has a distribution of 80%, and the testing data has a distribution of 20%.
Files
Steps to reproduce
We use the Python language. How to visualize the data with (.pcd) format? You can follow this code: import open3d as o3d import sys np.set_printoptions(threshold=sys.maxsize) pcd = o3d.io.read_point_cloud("D:/Farah/sqt000004.pcd") #NameOfFile How to convert PCAP to PCD, You can follow this code: from helper import * from pathlib import Path for folder in Path('C:/Users/Farah/PycharmProjects/coba/DataOri2/').iterdir(): tmp = Path(folder).stem for subfolder in Path(folder).rglob('*.json'): meta_file = Path(subfolder).stem f = os.path.join(folder, meta_file) meta_path = f+".json" pcap_path = f+".pcap" with open(meta_path, 'r') as f: info = client.SensorInfo(f.read()) source = pcap.Pcap(pcap_path, info) meta = source.metadata scans = client.Scans(source) if not os.path.exists("C:/Users/Farah/PycharmProjects/coba/DataOri2/PCD/"): os.makedirs("C:/Users/Farah/PycharmProjects/coba/DataOri2/PCD/") if not os.path.exists("C:/Users/Farah/PycharmProjects/coba/DataOri2/PCD/"): os.makedirs("C:/Users/Farah/PycharmProjects/coba/DataOri2/PCD/") if not os.path.exists("C:/Users/Farah/PycharmProjects/coba/DataOri2/PCD/pose"+tmp): os.makedirs("C:/Users/Farah/PycharmProjects/coba/DataOri2/PCD/pose"+tmp) write_path = "C:/Users/Farah/PycharmProjects/coba/DataOri2/PCD/pose"+str(tmp) print("Writing PCD to: ",write_path) pcap_to_pcd(source=source,metadata=meta, pcd_base=write_path+"/obj") #directory and file names can be customized
Institutions
Categories
Funding
Ministry of Education, Culture, Research, and Technology