Max Robotics — Global Service Robot Directory + US Certification

About

Siqi Shen is a leading researcher in computer vision and robotics, specializing in human-centric 4D scene capture, LiDAR localization, and cross-domain 3D perception. Their groundbreaking work, *HSC4D: Human-centered 4D Scene Capture in Large-scale Indoor-outdoor Space Using Wearable IMUs and LiDAR* (2022, 24 citations), introduces a novel method to create dynamic digital worlds by capturing human motions and environment interactions using only body-mounted sensors—eliminating the need for external cameras. This innovation enables accurate, space-free reconstruction of complex scenes, with applications in AR/VR and autonomous systems. Shen also advances efficient LiDAR localization with *LightLoc* (2025, 3 citations), reducing training time from days to minutes for time-critical autonomous driving upgrades. Their work on *Text to Point Cloud Localization* (2025, 3 citations) pioneers language-based spatial understanding, using multi-level negative contrastive learning to resolve ambiguity in text-scene matching. Additionally, Shen’s research on cross-domain descriptors (2021, 3 citations) enhances 2D-3D matching through hard triplet loss and spatial transformer networks, improving robustness in real-world environments. With a focus on scalable, sensor-efficient solutions, Shen’s contributions are shaping the future of embodied AI and spatial intelligence.