4 papers
FreeView: Robust 3D Reconstruction from Uncurated Views
Junho Kim, Hyunyoung Choi, Seongwon Lee
Reconstructing a coherent 3D scene from uncurated image collections remains challenging due to the presence of irrelevant or non-overlapping views. Recent work has shown that feed-…
RayOcc: Occlusion-Aware Ray Occupancy Estimation via Gaussian Mixture Intensity
Junho Kim, Seongwon Lee
Camera-only 3D semantic occupancy prediction aims to infer voxel-wise scene semantics from multi-view images, yet remains fundamentally challenging due to depth ambiguity and occlu…
VG3T: Visual Geometry Grounded Gaussian Transformer
Junho Kim, Seongwon Lee
Generating a coherent 3D scene representation from multi-view images is a fundamental yet challenging task. Existing methods often struggle with multi-view fusion, leading to fragm…
Lazy-DaSH: Lazy Approach for Hypergraph-based Multi-robot Task and Motion Planning
Seongwon Lee, James Motes, Isaac Ngui +2
We introduce Lazy-DaSH, an improvement over the recent state of the art multi-robot task and motion planning method DaSH, which scales to more than double the number of robots and…