Reproduce “Sixth-Sense: Self-Supervised Learning of Spatial Awareness of Humans from a Planar Lidar” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Simone Arreghini, Nicholas Carlotti, Mirko Nava, Antonio Paolillo, Alessandro Giusti
Sixth-Sense trains a model to estimate people’s presence, position, and relative orientation around a robot from planar-LiDAR scan history, using RGB-D camera detections as labels. On the Lab test set, 30 scans over three seconds improve P80 precision and orientation error over one scan. Distance error is slightly worse, and public-space evaluation is limited to camera-visible regions.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.