arXiv subjects

Zhijian Qiao

Publications and source records attributed to Zhijian Qiao.

16 recordsLinked to original sources

arXiv2026

MeRoPE: Metric Rotary Position Embedding for Camera-Controlled Video Generation

In camera-controlled video generation, geometry-aware positional encodings condition tokens on camera extrinsics and per-token viewing rays. Existing schemes, however, have a scale-dependent failure mode on real-world metric camera trajectories: homogeneous projective encodings cause attention logits and feature norms to grow unbounded with physical translation baselines. We propose MeRoPE (Metric Rotary Position Embedding), a norm-preserving relative camera encoding for attention. MeRoPE encodes relative orientations between calibrated viewing rays with orthogonal rotation blocks, maps raw metric displacements into multi-frequency rotary phases, and adds a disparity-anchored correspondence prior along the epipolar arc. This design strictly preserves feature norms, bounds pre-softmax attention logits regardless of the physical translation scale, and maintains exact invariance to global rigid coordinate changes. Across nuScenes and PanShot, which cover large-baseline trajectories and diverse camera optics, respectively, MeRoPE achieves stronger camera control than prior encodings, with the best consistency between generated camera motion and conditioning poses in both rotation and translation. Code will be made publicly available.

cs.CV↗

arXiv2025

CSMapping: Scalable Crowdsourced Semantic Mapping and Topology Inference for Autonomous Driving

Crowdsourcing enables scalable autonomous driving map construction, but low-cost sensor noise hinders quality from improving with data volume. We propose CSMapping, a system that produces accurate semantic maps and topological road centerlines whose quality consistently increases with more crowdsourced data. For semantic mapping, we train a latent diffusion model on HD maps (optionally conditioned on SD maps) to learn a generative prior of real-world map structure, without requiring paired crowdsourced/HD-map supervision. This prior is incorporated via constrained MAP optimization in latent space, ensuring robustness to severe noise and plausible completion in unobserved areas. Initialization uses a robust vectorized mapping module followed by diffusion inversion; optimization employs efficient Gaussian-basis reparameterization, projected gradient descent zobracket multi-start, and latent-space factor-graph for global consistency. For topological mapping, we apply confidence-weighted k-medoids clustering and kinematic refinement to trajectories, yielding smooth, human-like centerlines robust to trajectory variation. Experiments on nuScenes, Argoverse 2, and a large proprietary dataset achieve state-of-the-art semantic and topological mapping performance, with thorough ablation and scalability studies.

cs.CV↗

arXiv2025

SG-Reg: Generalizable and Efficient Scene Graph Registration

This paper addresses the challenges of registering two rigid semantic scene graphs, an essential capability when an autonomous agent needs to register its map against a remote agent, or against a prior map. The hand-crafted descriptors in classical semantic-aided registration, or the ground-truth annotation reliance in learning-based scene graph registration, impede their application in practical real-world environments. To address the challenges, we design a scene graph network to encode multiple modalities of semantic nodes: open-set semantic feature, local topology with spatial awareness, and shape feature. These modalities are fused to create compact semantic node features. The matching layers then search for correspondences in a coarse-to-fine manner. In the back-end, we employ a robust pose estimator to decide transformation according to the correspondences. We manage to maintain a sparse and hierarchical scene representation. Our approach demands fewer GPU resources and fewer communication bandwidth in multi-agent tasks. Moreover, we design a new data generation approach using vision foundation models and a semantic mapping module to reconstruct semantic scene graphs. It differs significantly from previous works, which rely on ground-truth semantic annotations to generate data. We validate our method in a two-agent SLAM benchmark. It significantly outperforms the hand-crafted baseline in terms of registration success rate. Compared to visual loop closure networks, our method achieves a slightly higher registration recall while requiring only 52 KB of communication bandwidth for each query frame. Code available at: \href{http://github.com/HKUST-Aerial-Robotics/SG-Reg}{http://github.com/HKUST-Aerial-Robotics/SG-Reg}.

cs.RO↗

arXiv2025

SLIM: Scalable and Lightweight LiDAR Mapping in Urban Environments

LiDAR point cloud maps are extensively utilized on roads for robot navigation due to their high consistency. However, dense point clouds face challenges of high memory consumption and reduced maintainability for long-term operations. In this study, we introduce SLIM, a scalable and lightweight mapping system for long-term LiDAR mapping in urban environments. The system begins by parameterizing structural point clouds into lines and planes. These lightweight and structural representations meet the requirements of map merging, pose graph optimization, and bundle adjustment, ensuring incremental management and local consistency. For long-term operations, a map-centric nonlinear factor recovery method is designed to sparsify poses while preserving mapping accuracy. We validate the SLIM system with multi-session real-world LiDAR data from classical LiDAR mapping datasets, including KITTI, NCLT, HeLiPR and M2DGR. The experiments demonstrate its capabilities in mapping accuracy, lightweightness, and scalability. Map re-use is also verified through map-based robot localization. Finally, with multi-session LiDAR data, the SLIM system provides a globally consistent map with low memory consumption (~130 KB/km on KITTI).

cs.RO↗

arXiv2025

Speak the Same Language: Global LiDAR Registration on BIM Using Pose Hough Transform

Light detection and ranging (LiDAR) point clouds and building information modeling (BIM) represent two distinct data modalities in the fields of robot perception and construction. These modalities originate from different sources and are associated with unique reference frames. The primary goal of this study is to align these modalities within a shared reference frame using a global registration approach, effectively enabling them to ``speak the same language''. To achieve this, we propose a cross-modality registration method, spanning from the front end to the back end. At the front end, we extract triangle descriptors by identifying walls and intersected corners, enabling the matching of corner triplets with a complexity independent of the BIM's size. For the back-end transformation estimation, we utilize the Hough transform to map the matched triplets to the transformation space and introduce a hierarchical voting mechanism to hypothesize multiple pose candidates. The final transformation is then verified using our designed occupancy-aware scoring method. To assess the effectiveness of our approach, we conducted real-world multi-session experiments in a large-scale university building, employing two different types of LiDAR sensors. We make the collected datasets and codes publicly available to benefit the community.

cs.RO↗

arXiv2025

SLABIM: A SLAM-BIM Coupled Dataset in HKUST Main Building

Existing indoor SLAM datasets primarily focus on robot sensing, often lacking building architectures. To address this gap, we design and construct the first dataset to couple the SLAM and BIM, named SLABIM. This dataset provides BIM and SLAM-oriented sensor data, both modeling a university building at HKUST. The as-designed BIM is decomposed and converted for ease of use. We employ a multi-sensor suite for multi-session data collection and mapping to obtain the as-built model. All the related data are timestamped and organized, enabling users to deploy and test effectively. Furthermore, we deploy advanced methods and report the experimental results on three tasks: registration, localization and semantic mapping, demonstrating the effectiveness and practicality of SLABIM. We make our dataset open-source at https://github.com/HKUST-Aerial-Robotics/SLABIM.

cs.RO↗

arXiv2024

FM-Fusion: Instance-aware Semantic Mapping Boosted by Vision-Language Foundation Models

Semantic mapping based on the supervised object detectors is sensitive to image distribution. In real-world environments, the object detection and segmentation performance can lead to a major drop, preventing the use of semantic mapping in a wider domain. On the other hand, the development of vision-language foundation models demonstrates a strong zero-shot transferability across data distribution. It provides an opportunity to construct generalizable instance-aware semantic maps. Hence, this work explores how to boost instance-aware semantic mapping from object detection generated from foundation models. We propose a probabilistic label fusion method to predict close-set semantic classes from open-set label measurements. An instance refinement module merges the over-segmented instances caused by inconsistent segmentation. We integrate all the modules into a unified semantic mapping system. Reading a sequence of RGB-D input, our work incrementally reconstructs an instance-aware semantic map. We evaluate the zero-shot performance of our method in ScanNet and SceneNN datasets. Our method achieves 40.3 mean average precision (mAP) on the ScanNet semantic instance segmentation task. It outperforms the traditional semantic mapping method significantly.

cs.CV↗

arXiv2024

G3Reg: Pyramid Graph-based Global Registration using Gaussian Ellipsoid Model

This study introduces a novel framework, G3Reg, for fast and robust global registration of LiDAR point clouds. In contrast to conventional complex keypoints and descriptors, we extract fundamental geometric primitives, including planes, clusters, and lines (PCL) from the raw point cloud to obtain low-level semantic segments. Each segment is represented as a unified Gaussian Ellipsoid Model (GEM), using a probability ellipsoid to ensure the ground truth centers are encompassed with a certain degree of probability. Utilizing these GEMs, we present a distrust-and-verify scheme based on a Pyramid Compatibility Graph for Global Registration (PAGOR). Specifically, we establish an upper bound, which can be traversed based on the confidence level for compatibility testing to construct the pyramid graph. Then, we solve multiple maximum cliques (MAC) for each level of the pyramid graph, thus generating the corresponding transformation candidates. In the verification phase, we adopt a precise and efficient metric for point cloud alignment quality, founded on geometric primitives, to identify the optimal candidate. The algorithm's performance is validated on three publicly available datasets and a self-collected multi-session dataset. Parameter settings remained unchanged during the experiment evaluations. The results exhibit superior robustness and real-time performance of the G3Reg framework compared to state-of-the-art methods. Furthermore, we demonstrate the potential for integrating individual GEM and PAGOR components into other registration frameworks to enhance their efficacy. Code: https://github.com/HKUST-Aerial-Robotics/G3Reg

cs.CV↗

arXiv2024

Less is More: Physical-enhanced Radar-Inertial Odometry

Radar offers the advantage of providing additional physical properties related to observed objects. In this study, we design a physical-enhanced radar-inertial odometry system that capitalizes on the Doppler velocities and radar cross-section information. The filter for static radar points, correspondence estimation, and residual functions are all strengthened by integrating the physical properties. We conduct experiments on both public datasets and our self-collected data, with different mobile platforms and sensor types. Our quantitative results demonstrate that the proposed radar-inertial odometry system outperforms alternative methods using the physical-enhanced components. Our findings also reveal that using the physical properties results in fewer radar points for odometry estimation, but the performance is still guaranteed and even improved, thus aligning with the ``less is more'' principle.

cs.RO↗

arXiv2023

Pyramid Semantic Graph-based Global Point Cloud Registration with Low Overlap

Global point cloud registration is essential in many robotics tasks like loop closing and relocalization. Unfortunately, the registration often suffers from the low overlap between point clouds, a frequent occurrence in practical applications due to occlusion and viewpoint change. In this paper, we propose a graph-theoretic framework to address the problem of global point cloud registration with low overlap. To this end, we construct a consistency graph to facilitate robust data association and employ graduated non-convexity (GNC) for reliable pose estimation, following the state-of-the-art (SoTA) methods. Unlike previous approaches, we use semantic cues to scale down the dense point clouds, thus reducing the problem size. Moreover, we address the ambiguity arising from the consistency threshold by constructing a pyramid graph with multi-level consistency thresholds. Then we propose a cascaded gradient ascend method to solve the resulting densest clique problem and obtain multiple pose candidates for every consistency threshold. Finally, fast geometric verification is employed to select the optimal estimation from multiple pose candidates. Our experiments, conducted on a self-collected indoor dataset and the public KITTI dataset, demonstrate that our method achieves the highest success rate despite the low overlap of point clouds and low semantic quality. We have open-sourced our code https://github.com/HKUST-Aerial-Robotics/Pagor for this project.

cs.RO↗

arXiv2023

Online Monocular Lane Mapping Using Catmull-Rom Spline

In this study, we introduce an online monocular lane mapping approach that solely relies on a single camera and odometry for generating spline-based maps. Our proposed technique models the lane association process as an assignment issue utilizing a bipartite graph, and assigns weights to the edges by incorporating Chamfer distance, pose uncertainty, and lateral sequence consistency. Furthermore, we meticulously design control point initialization, spline parameterization, and optimization to progressively create, expand, and refine splines. In contrast to prior research that assessed performance using self-constructed datasets, our experiments are conducted on the openly accessible OpenLane dataset. The experimental outcomes reveal that our suggested approach enhances lane association and odometry precision, as well as overall lane map quality. We have open-sourced our code1 for this project.

cs.RO↗

arXiv2023

SeasonDepth: Cross-Season Monocular Depth Prediction Dataset and Benchmark under Multiple Environments

Different environments pose a great challenge to the outdoor robust visual perception for long-term autonomous driving, and the generalization of learning-based algorithms on different environments is still an open problem. Although monocular depth prediction has been well studied recently, few works focus on the robustness of learning-based depth prediction across different environments, e.g. changing illumination and seasons, owing to the lack of such a multi-environment real-world dataset and benchmark. To this end, the first cross-season monocular depth prediction dataset and benchmark, SeasonDepth, is introduced to benchmark the depth estimation performance under different environments. We investigate several state-of-the-art representative open-source supervised and self-supervised depth prediction methods using newly-formulated metrics. Through extensive experimental evaluation on the proposed dataset and cross-dataset evaluation with current autonomous driving datasets, the performance and robustness against the influence of multiple environments are analyzed qualitatively and quantitatively. We show that long-term monocular depth prediction is still challenging and believe our work can boost further research on the long-term robustness and generalization for outdoor visual perception. The dataset is available on https://seasondepth.github.io, and the benchmark toolkit is available on https://github.com/ SeasonDepth/SeasonDepth.

cs.CV↗

arXiv2023

Multi-Session, Localization-oriented and Lightweight LiDAR Mapping Using Semantic Lines and Planes

In this paper, we present a centralized framework for multi-session LiDAR mapping in urban environments, by utilizing lightweight line and plane map representations instead of widely used point clouds. The proposed framework achieves consistent mapping in a coarse-to-fine manner. Global place recognition is achieved by associating lines and planes on the Grassmannian manifold, followed by an outlier rejection-aided pose graph optimization for map merging. Then a novel bundle adjustment is also designed to improve the local consistency of lines and planes. In the experimental section, both public and self-collected datasets are used to demonstrate efficiency and effectiveness. Extensive results validate that our LiDAR mapping framework could merge multi-session maps globally, optimize maps incrementally, and is applicable for lightweight robot localization.

cs.RO↗

arXiv2021

A Registration-aided Domain Adaptation Network for 3D Point Cloud Based Place Recognition

In the field of large-scale SLAM for autonomous driving and mobile robotics, 3D point cloud based place recognition has aroused significant research interest due to its robustness to changing environments with drastic daytime and weather variance. However, it is time-consuming and effort-costly to obtain high-quality point cloud data for place recognition model training and ground truth for registration in the real world. To this end, a novel registration-aided 3D domain adaptation network for point cloud based place recognition is proposed. A structure-aware registration network is introduced to help to learn features with geometric information and a 6-DoFs pose between two point clouds with partial overlap can be estimated. The model is trained through a synthetic virtual LiDAR dataset through GTA-V with diverse weather and daytime conditions and domain adaptation is implemented to the real-world domain by aligning the global features. Our results outperform state-of-the-art 3D place recognition baselines or achieve comparable on the real-world Oxford RobotCar dataset with the visualization of registration on the virtual dataset.

cs.CV↗

arXiv2021

End-to-End 3D Point Cloud Learning for Registration Task Using Virtual Correspondences

3D Point cloud registration is still a very challenging topic due to the difficulty in finding the rigid transformation between two point clouds with partial correspondences, and it's even harder in the absence of any initial estimation information. In this paper, we present an end-to-end deep-learning based approach to resolve the point cloud registration problem. Firstly, the revised LPD-Net is introduced to extract features and aggregate them with the graph network. Secondly, the self-attention mechanism is utilized to enhance the structure information in the point cloud and the cross-attention mechanism is designed to enhance the corresponding information between the two input point clouds. Based on which, the virtual corresponding points can be generated by a soft pointer based method, and finally, the point cloud registration problem can be solved by implementing the SVD method. Comparison results in ModelNet40 dataset validate that the proposed approach reaches the state-of-the-art in point cloud registration tasks and experiment resutls in KITTI dataset validate the effectiveness of the proposed approach in real applications.Our source code is available at \url{https://github.com/qiaozhijian/VCR-Net.git}

cs.CV↗

arXiv2020

DASGIL: Domain Adaptation for Semantic and Geometric-aware Image-based Localization

Long-Term visual localization under changing environments is a challenging problem in autonomous driving and mobile robotics due to season, illumination variance, etc. Image retrieval for localization is an efficient and effective solution to the problem. In this paper, we propose a novel multi-task architecture to fuse the geometric and semantic information into the multi-scale latent embedding representation for visual place recognition. To use the high-quality ground truths without any human effort, the effective multi-scale feature discriminator is proposed for adversarial training to achieve the domain adaptation from synthetic virtual KITTI dataset to real-world KITTI dataset. The proposed approach is validated on the Extended CMU-Seasons dataset and Oxford RobotCar dataset through a series of crucial comparison experiments, where our performance outperforms state-of-the-art baselines for retrieval-based localization and large-scale place recognition under the challenging environment.

cs.CV↗