# CVSD-Reg Distills DINOv2 Semantics for Zero-Shot Cross-Sensor LiDAR Registration

- Source: 3D视觉工坊 · 微信公众号
- Published: 2026-10-20T23:00:00.000Z
- RoboHot: https://robohot.news/en/items/g2m3zw7ythx6ls6ohs4byeehb
- Original: https://mp.weixin.qq.com/s?__biz=MzU1MjY4MTA1MQ==&mid=2247752400&idx=2&sn=f82ce9b5c99869d4a6d4a21d1dbbd7ca&chksm=fab35fa83117f924d51c7ab7b0e03bd8fc61e185a898d88b1f54296800aef8ce6cfe5d0ced4c&scene=126&sessionid=1792676434#rd

## Summary

CVSD-Reg is a robust global LiDAR registration framework that distills frozen DINOv2 visual semantic priors into a Point Transformer V3 student network, achieving zero-shot cross-sensor generalization with no camera needed at inference time.