Related Experiment Video
Updated: Jan 16, 2026
08:16
Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring
Published on: October 24, 2025
488
3D human pose point cloud data of light detection and ranging (LiDAR)
Farah Zakiyah Rahmanti1,2, Moch Iskandar Riansyah1,3, Oddy Virgantara Putra4
1Department of Electrical Engineering, Institut Teknologi Sepuluh Nopember, Surabaya, 60111, Indonesia.
Data in Brief
|September 29, 2025
Summary
This study introduces a new 3D human pose dataset using 3D Light Detection and Ranging (LiDAR) for privacy-preserving pose prediction. The dataset, comprising 1400 point cloud samples, enables deep learning models like CNNs for accurate human pose recognition.
Area of Science:
- Computer Vision
- Deep Learning
- Robotics
- Human-Computer Interaction
Background:
- 3D Light Detection and Ranging (LiDAR) sensors are integral to autonomous systems, commonly used for object detection in vehicles.
- LiDAR's ability to generate 3D point clouds offers a privacy-friendly alternative to cameras for indoor human pose estimation.
- Existing methods often rely on visual data, posing privacy concerns, whereas LiDAR captures spatial and temporal data without facial recognition.
Purpose of the Study:
- To develop and present a novel 3D point cloud dataset for human pose prediction using 3D LiDAR technology.
- To facilitate the advancement of deep learning models for accurate and privacy-conscious human pose recognition.
- To provide a standardized dataset for training and testing algorithms in human-computer interaction and robotics.
Main Methods:
- Collected 1400 3D point cloud samples of four human poses (hands-to-the-side, sit-down, squat-down, stand-up) using an Ouster OS1 3D LiDAR sensor.
- Processed raw spatio-temporal data from PCAP and JSON files into PCD format, creating a dataset of 280 samples per pose class for training and testing.
- Utilized a tripod-mounted 3D LiDAR at a distance of 120 cm indoors, with data collection occurring between 10 a.m. and 1 p.m.
Main Results:
- A comprehensive 3D human pose dataset in PCD format, totaling 1400 samples across four distinct pose classes.
- The dataset is suitable for training deep learning models, including Convolutional Neural Networks (CNNs), for human pose prediction.
- Demonstrated the feasibility of using 3D LiDAR for generating rich spatio-temporal data for human activity analysis.
Conclusions:
- The developed 3D LiDAR human pose dataset offers a privacy-preserving approach for pose prediction and activity recognition.
- This dataset serves as a valuable resource for researchers in computer vision and deep learning to develop advanced human pose estimation algorithms.
- Future work can expand the dataset by including diverse demographics and more complex human activities to enhance model generalizability.
Related Concept Videos
01:30Three-Dimensional Force System
2.8K
In mechanical engineering, a three-dimensional force system is a system of forces acting in three dimensions, with forces applied along the x, y, and z coordinate axes. The three-dimensional force system is an important concept in mechanical engineering, as it allows engineers to understand and analyze the behavior of objects and structures in three dimensions. By understanding the forces acting on a system, engineers can design more efficient and effective mechanical systems that can withstand...
2.8K
02:16Light Acquisition
9.4K
In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.
9.4K