5 papers
Monocular Visual Place Recognition in LiDAR Maps via Cross-Modal State Space Model and Multi-View Matching
Gongxin Yao, Xinyang Li, Luowei Fu +1
Achieving monocular camera localization within pre-built LiDAR maps can bypass the simultaneous mapping process of visual SLAM systems, potentially reducing the computational overh…
CMR-Agent: Learning a Cross-Modal Agent for Iterative Image-to-Point Cloud Registration
Gongxin Yao, Yixin Xuan, Xinyang Li +1
Image-to-point cloud registration aims to determine the relative camera pose of an RGB image with respect to a point cloud. It plays an important role in camera localization within…
MaFreeI2P: A Matching-Free Image-to-Point Cloud Registration Paradigm with Active Camera Pose Retrieval
Gongxin Yao, Xinyang Li, Yixin Xuan +1
Image-to-point cloud registration seeks to estimate their relative camera pose, which remains an open question due to the data modality gaps. The recent matching-based methods tend…
FAGhead: Fully Animate Gaussian Head from Monocular Videos
Yixin Xuan, Xinyang Li, Gongxin Yao +4
High-fidelity reconstruction of 3D human avatars has a wild application in visual reality. In this paper, we introduce FAGhead, a method that enables fully controllable human portr…
GGAvatar: Geometric Adjustment of Gaussian Head Avatar
Xinyang Li, Jiaxin Wang, Yixin Xuan +2
We propose GGAvatar, a novel 3D avatar representation designed to robustly model dynamic head avatars with complex identities and deformations. GGAvatar employs a coarse-to-fine st…