With the recent growth of urban mapping and autonomous driving efforts, there has been an
explosion of raw 3D data collected from terrestrial platforms with lidar scanners and color …

Open-vocabulary 3D instance segmentation is cutting-edge for its ability to segment 3D
instances without predefined categories. However progress in 3D lags behind its 2D …

While recent 3D instance segmentation approaches show promising results based on
transformer architectures they often fail to correctly identify instances with similar …

Recent advances in 3D semantic scene understanding have shown impressive progress in
3D instance segmentation, enabling object-level reasoning about 3D scenes; however, a …

Semantic annotations are indispensable to train or evaluate perception models, yet very
costly to acquire. This work introduces a fully automated 2D/3D labeling framework that …

Learning unknown objects in the environment is important for detection and manipulation
tasks. Prior to learning the unknown objects the ground-truth labels have to be provided. The …

Abstract 3D instance segmentation (3DIS) is a crucial task but point-level annotations are
tedious in fully supervised settings. Thus using bounding boxes (bboxes) as annotations has …

We introduce 3D-SIS, a novel neural network architecture for 3D semantic instance
segmentation in commodity RGB-D scans. The core idea of our method to jointly learn from …

Recent advancements in deep learning techniques have accelerated the growth of robotic
vision systems. One way this technology can be applied is to use a mobile robot to …

Inferring detailed 3D geometry of the scene is crucial for robotics applications, simulation,
and 3D content creation. However, such information is hard to obtain, and thus very few …