In this paper, we present a neural sampling method, EllipssianNet, which predicts 2D Gaussians from an input RGB image. Trained with a Voronoi diagram-based synthetic dataset, EllipssianNet outputs a center map and a covariance map, which are combined with the colors sampled from the input image to generate 2D Gaussians. The Gaussians are anisotropic and aligned with local complexities of the input RGB image. The 2D Gaussians are converted into 3D ones that are then optimized and rasterized in the 3D Gaussian Splatting framework. EllipssianNet is tested in two applications. In Gaussian-based image representation, initialization with EllipssianNet enables faster convergence and higher rendering quality. EllipssianNet is also seamlessly integrated into a real-time SLAM system, producing high-quality reconstructions under online constraints. The code and trained weight are available at https://github.com/medialab-ku/EllipssianNet

Abstract This paper presents a novel method that automatically reconstructs pipes in real time using an RGB‐D camera. The input image is decomposed into superpixels, and a pipe element, which is a 3D circle lying on the pipe surface, is generated for each superpixel. Over frames, the pipe elements are grouped into sequences and are finally modeled in parametric equations. The method is tested in daily settings, where the pipes may be curved and their radii may vary along the axes. The test results show that the method is able to reconstruct pipes efficiently, precisely and robustly.

Recent advances in 3D reconstruction and tracking technologies have made it possible to volumetrically capture human body and performance at real time. In the field of human-computer interaction, however, no works have been reported on the user study made with such volumetric capture avatars. This paper investigates how the volumetric capture avatar affects users' sense of social presence in immersive virtual environments. In our experiments, the volumetric capture avatar of an actor is compared with the actor captured in 2D video and another 3D avatar obtained by pre-scanning the actor. The experiment results show that users have the highest sense of social presence with the volumetric capture avatar when performing dynamic tasks whereas they have higher sense of social presence with the volumetric capture avatar and 2D video than with the pre-scanned avatar when performing static tasks. These imply that the emerging volumetric capture techniques can be an attractive tool for mixed reality, telepresence, and many other 3D applications.