Imaging Hidden Objects with Consumer LiDAR via Motion Induced Sampling
English
United States, San Francisco
Massachusetts Institute of Technology
数据描述

This dataset centers on smartphone-grade and other consumer LiDAR measurements repurposed for non-line-of-sight imaging, capturing indirect light returns that reveal objects hidden around corners. It contains data and associated reconstructions for hidden-object 3D shape recovery, real-time tracking, camera localization, and hand tracking using off-the-shelf LiDAR sensors. Its purpose is to demonstrate that consumer depth hardware in phones, headsets, robots, and vehicles can function as practical plug-and-play around-the-corner sensing systems without new hardware. The dataset supports applications in robotics, autonomous driving, augmented reality, indoor localization, scene reconstruction, and search-and-rescue.
cornar.media.mit.edu
IP: 185.199.109.153
加载中...
相关论文
1Imaging hidden objects with consumer LiDAR via motion-induced sampling
Siddharth Somasundaram•Aaron Young•Akshat Dave•Adithya Pediredla•Ramesh Raskar
Nature
2026
•2026/5/20
•Vol.653 No.8115 p.693-699
Light-detection and ranging (LiDAR) is being increasingly deployed for consumer imaging across handheld, wearable and robotic applications1–4. These sensors measure the time-of-flight of light at picosecond resolution, which could enable them to image objects hidden from their field of view. Althoug...
Applied opticsEngineeringImaging and sensingInformation theory and computation
10.1038/S41586-026-10502-X
ISSN:0028-0836