arXiv subjects

Youngji Kim

Publications and source records attributed to Youngji Kim.

5 recordsLinked to original sources

arXiv2026

InSpace: Structure-Aware 3D Indoor Scene Generation from a Single 360{\deg} Image

Recent advances in single image-to-3D generation have enabled high-quality asset synthesis, yet extending these capabilities to indoor scene generation remains challenging. Existing methods focus on asset-level generation while neglecting the structural layout, which is essential for downstream applications and serves as the spatial anchor for grounding assets. However, a single image with a limited field of view lacks the spatial coverage to recover a coherent global layout. To this end, we use a 360{\deg} image represented in equirectangular projection (ERP) and propose InSpace, a structure-aware framework for 3D indoor scene generation. InSpace comprises three stages: (1) estimating partial scene geometry as spatial priors, (2) generating coarse scene structure with view-selective cross-attention, and (3) producing detailed layout and asset geometry with textures through a global-local hybrid attention, using flow matching. We also propose ERP-FRONT, a paired ERP-Image-to-3D indoor scene dataset based on 3D-FRONT. Experiments show that InSpace generates complete 3D indoor scenes with structural layout, along with separate textured assets from a single ERP image, achieving strong performance across 3D and 2D metrics. Project Page: https://kookie12.github.io/InSpace-Project-Page/

cs.CV↗

arXiv2026

Tailoring phonon-driven responses in {\alpha}-MoO3 through isotopic enrichment

The implementation of polaritonic materials into nanoscale devices requires selective tuning of parameters to realize desired spectral or thermal responses. One robust material is {\alpha}-MoO3, which as an orthorhombic crystal boasts three distinct phonon dispersions, providing three polaritonic dispersions of hyperbolic phonon polaritons (HPhPs) across the mid-infrared (MIR). Here, the tunability of both optical and thermal responses in isotopically enriched {\alpha}-MoO3 (98MoO3, Mo18O3 and 98Mo18O3) are explored. A uniform ~5 % spectral redshift from 18O enrichment is observed in both Raman- and IR-active TO phonons. Both the in- and out-of-plane thermal conductivities for the isotopic variations are reported. Ab initio calculations both replicate experimental findings and analyze the select-mode three-phonon scattering contributions. The HPhPs from each isotopic variation are probed with s-SNOM and their Q- factors are reported. A Q-factor maxima increase of ~50 % along the [100] in the RB2 and ~100 % along the [001] in the RB3 are reported for HPhPs supported in 98Mo18O3. Observations in both real and Fourier space of higher-order HPhP modes propagating in single slabs of isotopically enriched {\alpha}-MoO3 without the use of a subdiffractional surface scatterer are presented here. This work illustrates the tunability of {\alpha}-MoO3 for thermal and nanophotonic applications.

cond-mat.mtrl-sci↗

arXiv2022

A Single Correspondence Is Enough: Robust Global Registration to Avoid Degeneracy in Urban Environments

Global registration using 3D point clouds is a crucial technology for mobile platforms to achieve localization or manage loop-closing situations. In recent years, numerous researchers have proposed global registration methods to address a large number of outlier correspondences. Unfortunately, the degeneracy problem, which represents the phenomenon in which the number of estimated inliers becomes lower than three, is still potentially inevitable. To tackle the problem, a degeneracy-robust decoupling-based global registration method is proposed, called Quatro. In particular, our method employs quasi-SO(3) estimation by leveraging the Atlanta world assumption in urban environments to avoid degeneracy in rotation estimation. Thus, the minimum degree of freedom (DoF) of our method is reduced from three to one. As verified in indoor and outdoor 3D LiDAR datasets, our proposed method yields robust global registration performance compared with other global registration methods, even for distant point cloud pairs. Furthermore, the experimental results confirm the applicability of our method as a coarse alignment. Our code is available: https://github.com/url-kaist/quatro.

cs.CV↗

arXiv2021

Multitask Learning for Scalable and Dense Multilayer Bayesian Map Inference

This article presents a novel and flexible multitask multilayer Bayesian mapping framework with readily extendable attribute layers. The proposed framework goes beyond modern metric-semantic maps to provide even richer environmental information for robots in a single mapping formalism while exploiting intralayer and interlayer correlations. It removes the need for a robot to access and process information from many separate maps when performing a complex task, advancing the way robots interact with their environments. To this end, we design a multitask deep neural network with attention mechanisms as our front-end to provide heterogeneous observations for multiple map layers simultaneously. Our back-end runs a scalable closed-form Bayesian inference with only logarithmic time complexity. We apply the framework to build a dense robotic map including metric-semantic occupancy and traversability layers. Traversability ground truth labels are automatically generated from exteroceptive sensory data in a self-supervised manner. We present extensive experimental results on publicly available datasets and data collected by a 3D bipedal robot platform and show reliable mapping performance in different environments. Finally, we also discuss how the current framework can be extended to incorporate more information such as friction, signal strength, temperature, and physical quantity concentration using Gaussian map layers. The software for reproducing the presented results or running on customized data is made publicly available.

cs.RO↗

arXiv2019

Sequential Learning of Visual Tracking and Mapping Using Unsupervised Deep Neural Networks

We proposed an end-to-end deep learning-based simultaneous localization and mapping (SLAM) system following conventional visual odometry (VO) pipelines. The proposed method completes the SLAM framework by including tracking, mapping, and sequential optimization networks while training them in an unsupervised manner. Together with the camera pose and depth map, we estimated the observational uncertainty to make our system robust to noises such as dynamic objects. We evaluated our method using public indoor and outdoor datasets. The experiment demonstrated that our method works well in tracking and mapping tasks and performs comparably with other learning-based VO approaches. Notably, the proposed uncertainty modeling and sequential training yielded improved generality in a variety of environments.

cs.RO↗