6 papers
BudgetSplat: Budget-Controllable Generalizable Gaussian Splatting
Yuxi Hu, Jun Zhang, Wei Mao +2
Feed-forward 3D Gaussian Splatting enables efficient generalizable novel view synthesis, but existing pixel-aligned methods often generate highly redundant Gaussian primitives, cau…
CrossPressor: Scalable Cross-View Fusion for Feed-Forward 3D Gaussian Splatting
Yuxi Hu, Wenhao Dong, Jun Zhang +1
Feed-forward 3D Gaussian Splatting (3D-GS) becomes increasingly difficult to scale to long multi-view sequences. Anchor-based compression methods such as ZPressor alleviate represe…
A Height-Constrained 2-Point Minimal Solver for Pose Estimation from Active LED Markers with Event Cameras
Runze Yuan, Alexander Kappler, Jun Zhang +5
In many autonomous applications requiring real-time localization, active marker-based systems are preferred due to their low latency and ease of deployment compared to computationa…
-GS: Learning Context-aware, Cross-dimension, Cross-scale Feature for Generalizable Gaussian Splatting
Yuxi Hu, Jun Zhang, Kuangyi Chen +2
Generalizable Gaussian Splatting aims to synthesize novel views for unseen scenes without per-scene optimization. In particular, recent advancements utilize feed-forward networks t…
ICG-MVSNet: Learning Intra-view and Cross-view Relationships for Guidance in Multi-View Stereo
Yuxi Hu, Jun Zhang, Zhe Zhang +5
Multi-view Stereo (MVS) aims to estimate depth and reconstruct 3D point clouds from a series of overlapping images. Recent learning-based MVS frameworks overlook the geometric info…
EVLoc: Event-based Visual Localization in LiDAR Maps via Event-Depth Registration
Kuangyi Chen, Jun Zhang, Friedrich Fraundorfer
Event cameras are bio-inspired sensors with some notable features, including high dynamic range and low latency, which makes them exceptionally suitable for perception in challengi…