5 papers
Clutt3R-Seg: Sparse-view 3D Instance Segmentation for Language-grounded Grasping in Cluttered Scenes
Jeongho Noh, Tai Hyoung Rhee, Eunho Lee +3
Reliable 3D instance segmentation is fundamental to language-grounded robotic manipulation. Its critical application lies in cluttered environments, where occlusions, limited viewp…
Informative Object-centric Next Best View for Object-aware 3D Gaussian Splatting in Cluttered Scenes
Seunghoon Jeong, Eunho Lee, Jeongyun Kim +1
In cluttered scenes with inevitable occlusions and incomplete observations, selecting informative viewpoints is essential for building a reliable representation. In this context, 3…
TRAN-D: 2D Gaussian Splatting-based Sparse-view Transparent Object Depth Reconstruction via Physics Simulation for Scene Update
Jeongyun Kim, Seunghoon Jeong, Giseop Kim +3
Understanding the 3D geometry of transparent objects from RGB images is challenging due to their inherent physical properties, such as reflection and refraction. To address these d…
TranSplat: Surface Embedding-guided 3D Gaussian Splatting for Transparent Object Manipulation
Jeongyun Kim, Jeongho Noh, Dong-Guw Lee +1
Transparent object manipulation remains a significant challenge in robotics due to the difficulty of acquiring accurate and dense depth measurements. Conventional depth sensors oft…
Thermal Chameleon: Task-Adaptive Tone-mapping for Radiometric Thermal-Infrared images
Dong-Guw Lee, Jeongyun Kim, Younggun Cho +1
Thermal Infrared (TIR) imaging provides robust perception for navigating in challenging outdoor environments but faces issues with poor texture and low image contrast due to its 14…