most citedWhat Is The Best 3D Scene Representation for Robotics? From Geometric to Foundation Models

1 citations · 1 across the 4 of their papers we have counts for

collaborators

6 papers

cs.RO20261 cited

What Is The Best 3D Scene Representation for Robotics? From Geometric to Foundation Models

Tianchen Deng, Yue Pan, Shenghai Yuan +10

In this paper, we provide a comprehensive overview of existing scene representation methods for robotics, covering traditional representations such as point clouds, voxels, signed…

cs.CV2026

ReCCur: A Recursive Corner-Case Curation Framework for Robust Vision-Language Understanding in Open and Edge Scenarios

Yihan Wei, Shenghai Yuan, Tianchen Deng +2

Corner cases are rare or extreme scenarios that drive real-world failures, but they are difficult to curate at scale: web data are noisy, labels are brittle, and edge deployments p…

cs.CV2025

Reloc-VGGT: Visual Re-localization with Geometry Grounded Transformer

Tianchen Deng, Wenhua Wu, Kunzhen Wu +7

Visual localization has traditionally been formulated as a pair-wise pose regression problem. Existing approaches mainly estimate relative poses between two images and employ a lat…

cs.CV2025

UNO: Unified Self-Supervised Monocular Odometry for Platform-Agnostic Deployment

Wentao Zhao, Yihe Niu, Yanbo Wang +5

This work presents UNO, a unified monocular visual odometry framework that enables robust and adaptable pose estimation across diverse environments, platforms, and motion patterns.…

cs.CV2025

MCN-SLAM: Multi-Agent Collaborative Neural SLAM with Hybrid Implicit Neural Scene Representation

Tianchen Deng, Guole Shen, Xun Chen +9

Neural implicit scene representations have recently shown promising results in dense visual SLAM. However, existing implicit SLAM algorithms are constrained to single-agent scenari…

cs.CV2025

VPGS-SLAM: Voxel-based Progressive 3D Gaussian SLAM in Large-Scale Scenes

Tianchen Deng, Wenhua Wu, Junjie He +4

3D Gaussian Splatting has recently shown promising results in dense visual SLAM. However, existing 3DGS-based SLAM methods are all constrained to small-room scenarios and struggle…