Hezi Shi

Papers

Total Citations

H-Index

About

Hezi Shi is a researcher focused on advancing indoor robot perception and navigation through the integration of deep learning and semantic mapping. His primary research areas include 3D object detection, environmental perception, and visual semantic navigation for service robots. Shi’s most cited work, "Multi-Channel Convolutional Neural Network Based 3D Object Detection for Indoor Robot Environmental Perception" (2019, 24 citations), introduces a novel approach that elevates robot perception beyond low-level geometric reconstruction by enabling semantic understanding of indoor environments. This contribution addresses a critical gap in robotics, allowing machines to interpret their surroundings more like humans. In his related work, "Object-Aware Hybrid Map for Indoor Robot Visual Semantic Navigation" (2019, 5 citations), Shi proposes an innovative mapping framework that combines object-level semantics with traditional metric maps, facilitating more intuitive human-robot interaction and navigation. These contributions demonstrate Shi’s impact in bridging perception and semantics for autonomous systems, laying groundwork for smarter, context-aware robots capable of operating effectively in complex indoor spaces.