12 papers
Does Robust VIO Need More Learning? Geometry-Verified Visual Measurements under Distribution Shift
Yangyang Ning, Shu Liang, Quanbo Ge +3
Learning is increasingly introduced into visual-inertial odometry (VIO), ranging from learned feature front-ends to learning-dominant motion and geometry estimation. However, learn…
FAM-HRI: Foundation-Model Assisted Multi-Modal Human-Robot Interaction Combining Gaze and Speech
Yuzhi Lai, Shenghai Yuan, Peizheng Li +4
ffective Human-Robot Interaction (HRI) is crucial for enhancing accessibility and usability in real-world robotics applications. However, existing solutions often rely on gesture-…
Compact 3D Gaussian Splatting For Dense Visual SLAM
Tianchen Deng, Chang Nie, Shuhong Liu +6
Recent work has shown that 3D Gaussian-based SLAM enables high-quality reconstruction, accurate pose estimation, and real-time rendering of scenes. However, these approaches are bu…
UniLGL: Learning Uniform Place Recognition for FOV-limited/Panoramic LiDAR Global Localization
Hongming Shen, Xun Chen, Yulin Hui +5
Existing LGL methods typically consider only partial information (e.g., geometric features) from LiDAR observations or are designed for homogeneous LiDAR sensors, overlooking the u…
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…
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…