Giseop Kim

Hyoseok Ju, Giseop Kim

Crowd-sourced cooperative mapping from monocular cameras promises scalable 3D reconstruction without specialized sensors, yet remains hindered by two scale-specific failure modes: abrupt scale collapse from false-positive loop closures in repetitive environments, and gradual scale drift over long trajectories and per-robot scale ambiguity that prevent direct multi-session fusion. We present MR$.$ScaleMaster, a cooperative mapping system for crowd-sourced monocular videos that addresses both failure modes. MR$.$ScaleMaster introduces three key mechanisms. First, a Scale Collapse Alarm rejects spurious loop closures before they corrupt the pose graph. Second, a Sim(3) anchor node formulation generalizes the classical SE(3) framework to explicitly estimate per-session scale, resolving per-robot scale ambiguity and enforcing global scale consistency. Third, a modular, open-source, plug-and-play interface enables any monocular reconstruction model to integrate without backend modification. On KITTI sequences with up to 15 agents, the Sim(3) formulation achieves a 7.2x ATE reduction over the SE(3) baseline, and the alarm rejects all false-positive loops while preserving every valid constraint. We further demonstrate heterogeneous multi-robot dense mapping fusing MASt3R-SLAM, pi3, and VGGT-SLAM 2.0 within a single unified map.

Minsu Kim, Giseop Kim, Kyong Hwan Jin et al.

A recent sensor fusion in a Bird's Eye View (BEV) space has shown its utility in various tasks such as 3D detection, map segmentation, etc. However, the approach struggles with inaccurate camera BEV estimation, and a perception of distant areas due to the sparsity of LiDAR points. In this paper, we propose a broad BEV fusion (BroadBEV) that addresses the problems with a spatial synchronization approach of cross-modality. Our strategy aims to enhance camera BEV estimation for a broad-sighted perception while simultaneously improving the completion of LiDAR's sparsity in the entire BEV space. Toward that end, we devise Point-scattering that scatters LiDAR BEV distribution to camera depth distribution. The method boosts the learning of depth estimation of the camera branch and induces accurate location of dense camera features in BEV space. For an effective BEV fusion between the spatially synchronized features, we suggest ColFusion that applies self-attention weights of LiDAR and camera BEV features to each other. Our extensive experiments demonstrate that BroadBEV provides a broad-sighted BEV perception with remarkable performance gains.

Jinwoo Lee, Jiwoo Kim, Woojae Shin et al.

Point clouds are a fundamental representation for robotic perception tasks such as localization, mapping, and object pose estimation. However, LiDAR-acquired point clouds are inherently sparse and non-uniform, providing incomplete observations of the underlying scene geometry. This makes reliable geometric reasoning challenging and degrades downstream perception performance. Existing approaches attempt to compensate for these limitations by estimating local geometry, but often rely on hand-crafted statistics or end-to-end supervised learning, which can suffer from limited scalability or require large amounts of accurately labeled data. To address these challenges, we explicitly model point cloud geometry under a principled mathematical formulation. We represent local geometry as a statistical manifold induced by a family of Gaussian distributions, where each point is associated with a Gaussian capturing its local geometric structure. Based on this formulation, we introduce Point-to-Ellipsoid (POLI), a deep neural estimator that predicts per-point Gaussian geometry. POLI learns a mapping from point cloud observations to their underlying geometry in a self-supervised manner, removing the need for labeled data while preserving strong geometric inductive biases. The resulting representation integrates seamlessly into existing robotic perception pipelines without architectural modifications. Extensive experiments show that POLI enables accurate and robust geometry estimation and consistently improves performance across diverse robotic perception tasks.

Giseop Kim, Seungsang Yun, Jeongyun Kim et al.

Accurate 3D point cloud map generation is a core task for various robot missions or even for data-driven urban analysis. To do so, light detection and ranging (LiDAR) sensor-based simultaneous localization and mapping (SLAM) technology have been elaborated. To compose a full SLAM system, many odometry and place recognition methods have independently been proposed in academia. However, they have hardly been integrated or too tightly combined so that exchanging (upgrading) either single odometry or place recognition module is very effort demanding. Recently, the performance of each module has been improved a lot, so it is necessary to build a SLAM system that can effectively integrate them and easily replace them with the latest one. In this paper, we release such a front-end agnostic LiDAR SLAM system, named SC-LiDAR-SLAM. We built a complete SLAM system by designing it modular, and successfully integrating it with Scan Context++ and diverse existing opensource LiDAR odometry methods to generate an accurate point cloud map

Giseop Kim, Sunwook Choi, Ayoung Kim

Place recognition is a key module in robotic navigation. The existing line of studies mostly focuses on visual place recognition to recognize previously visited places solely based on their appearance. In this paper, we address structural place recognition by recognizing a place based on structural appearance, namely from range sensors. Extending our previous work on a rotation invariant spatial descriptor, the proposed descriptor completes a generic descriptor robust to both rotation (heading) and translation when roll-pitch motions are not severe. We introduce two sub-descriptors and enable topological place retrieval followed by the 1-DOF semi-metric localization thereby bridging the gap between topological place retrieval and metric localization. The proposed method has been evaluated thoroughly in terms of environmental complexity and scale. The source code is available and can easily be integrated into existing LiDAR simultaneous localization and mapping (SLAM).

Giseop Kim, Ayoung Kim

Long-term 3D map management is a fundamental capability required by a robot to reliably navigate in the non-stationary real-world. This paper develops open-source, modular, and readily available LiDAR-based lifelong mapping for urban sites. This is achieved by dividing the problem into successive subproblems: multi-session SLAM (MSS), high/low dynamic change detection, and positive/negative change management. The proposed method leverages MSS and handles potential trajectory error; thus, good initial alignment is not required for change detection. Our change management scheme preserves efficacy in both memory and computation costs, providing automatic object segregation from a large-scale point cloud map. We verify the framework's reliability and applicability even under permanent year-level variation, through extensive real-world experiments with multiple temporal gaps (from day to year).

Minsu Kim, Dongyeun Lee, Jaemyung Yu et al.

Object detection is pivotal in computer vision, yet its immense computational demands make deployment slow and power-hungry, motivating quantization. However, task-irrelevant morphologies such as background clutter and sensor noise induce redundant activations (or anomalies). These anomalies expand activation ranges and skew activation distributions toward task-irrelevant responses, complicating bit allocation and weakening the preservation of informative features. Without a clear criterion to distinguish anomalies, suppressing them can inadvertently discard useful information. To address this, we present InlierQ, an inlier-centric post-training quantization approach that separates anomalies from informative inliers. InlierQ computes gradient-aware volume saliency scores, classifies each volume as an inlier or anomaly, and fits a posterior distribution over these scores using the Expectation-Maximization (EM) algorithm. This design suppresses anomalies while preserving informative features. InlierQ is label-free, drop-in, and requires only 64 calibration samples. Experiments on the COCO and nuScenes benchmarks show consistent reductions in quantization error for camera-based (2D and 3D) and LiDAR-based (3D) object detection.

Seokha Moon, Janghyun Baek, Giseop Kim et al.

3D occupancy prediction has emerged as a key perception task for autonomous driving, as it reconstructs 3D environments to provide a comprehensive scene understanding. Recent studies focus on integrating spatiotemporal information obtained from past observations to improve prediction accuracy, using a multi-frame fusion approach that processes multiple past frames together. However, these methods struggle with a trade-off between efficiency and accuracy, which significantly limits their practicality. To mitigate this trade-off, we propose StreamOcc, a novel framework that aggregates spatio-temporal information in a stream-based manner. StreamOcc consists of two key components: (i) Stream-based Voxel Aggregation, which effectively accumulates past observations while minimizing computational costs, and (ii) Query-guided Aggregation, which recurrently aggregates instance-level features of dynamic objects into corresponding voxel features, refining fine-grained details of dynamic objects. Experiments on the Occ3D-nuScenes dataset show that StreamOcc achieves state-of-the-art performance in real-time settings, while reducing memory usage by more than 50% compared to previous methods.

Jeongyun Kim, Seunghoon Jeong, Giseop Kim et al.

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 difficulties, especially in scenarios with sparse views and dynamic environments, we introduce TRAN-D, a novel 2D Gaussian Splatting-based depth reconstruction method for transparent objects. Our key insight lies in separating transparent objects from the background, enabling focused optimization of Gaussians corresponding to the object. We mitigate artifacts with an object-aware loss that places Gaussians in obscured regions, ensuring coverage of invisible surfaces while reducing overfitting. Furthermore, we incorporate a physics-based simulation that refines the reconstruction in just a few seconds, effectively handling object removal and chain-reaction movement of remaining objects without the need for rescanning. TRAN-D is evaluated on both synthetic and real-world sequences, and it consistently demonstrated robust improvements over existing GS-based state-of-the-art methods. In comparison with baselines, TRAN-D reduces the mean absolute error by over 39% for the synthetic TRansPose sequences. Furthermore, despite being updated using only one image, TRAN-D reaches a δ < 2.5 cm accuracy of 48.46%, over 1.5 times that of baselines, which uses six images. Code and more results are available at https://jeongyun0609.github.io/TRAN-D/.

Minsu Kim, Giseop Kim, Sunwook Choi

Recent advancements in Bird's Eye View (BEV) fusion for map construction have demonstrated remarkable mapping of urban environments. However, their deep and bulky architecture incurs substantial amounts of backpropagation memory and computing latency. Consequently, the problem poses an unavoidable bottleneck in constructing high-resolution (HR) BEV maps, as their large-sized features cause significant increases in costs including GPU memory consumption and computing latency, named diverging training costs issue. Affected by the problem, most existing methods adopt low-resolution (LR) BEV and struggle to estimate the precise locations of urban scene components like road lanes, and sidewalks. As the imprecision leads to risky motion planning like collision avoidance, the diverging training costs issue has to be resolved. In this paper, we address the issue with our novel BEVRestore mechanism. Specifically, our proposed model encodes the features of each sensor to LR BEV space and restores them to HR space to establish a memory-efficient map constructor. To this end, we introduce the BEV restoration strategy, which restores aliasing, and blocky artifacts of the up-scaled BEV features, and narrows down the width of the labels. Our extensive experiments show that the proposed mechanism provides a plug-and-play, memory-efficient pipeline, enabling an HR map construction with a broad BEV scope.

Younghun Cho, Giseop Kim, Sangmin Lee et al.

Using publicly accessible maps, we propose a novel vehicle localization method that can be applied without using prior light detection and ranging (LiDAR) maps. Our method generates OSM descriptors by calculating the distances to buildings from a location in OpenStreetMap at a regular angle, and LiDAR descriptors by calculating the shortest distances to building points from the current location at a regular angle. Comparing the OSM descriptors and LiDAR descriptors yields a highly accurate vehicle localization result. Compared to methods that use prior LiDAR maps, our method presents two main advantages: (1) vehicle localization is not limited to only places with previously acquired LiDAR maps, and (2) our method is comparable to LiDAR map-based methods, and especially outperforms the other methods with respect to the top one candidate at KITTI dataset sequence 00.

Younggun Cho, Giseop Kim, Ayoung Kim

Recently, learning-based ego-motion estimation approaches have drawn strong interest from studies mostly focusing on visual perception. These groundbreaking works focus on unsupervised learning for odometry estimation but mostly for visual sensors. Compared to images, a learning-based approach using Light Detection and Ranging (LiDAR) has been reported in a few studies where, most often, a supervised learning framework is proposed. In this paper, we propose a novel approach to geometry-aware deep LiDAR odometry trainable via both supervised and unsupervised frameworks. We incorporate the Iterated Closest Point (ICP) algorithm into a deep-learning framework and show the reliability of the proposed pipeline. We provide two loss functions that allow switching between supervised and unsupervised learning depending on the ground-truth validity in the training phase. An evaluation using the KITTI and Oxford RobotCar dataset demonstrates the prominent performance and efficiency of the proposed method when achieving pose accuracy.