Xueyang Kang

Ariel Herrera, Xueyang Kang, Atal Anil Kumar

Robotic manipulation of textiles remains challenging because continuous deformation and self-occlusions hinder the robust visual perception required to estimate the cloth's state. To address the lack of annotated real-world data, we developed a Blender-based synthetic pipeline exporting auto-annotated keypoints, and combined manually labeled renders with real-world data to train a wrinkle detector. We present a perception framework integrating a CNN for permutation-invariant keypoint detection and a YOLOv8-OpenCV pipeline to extract grasping points from structural wrinkles. A proposed bimanual algorithm uses this system to stretch fully folded garments via wrinkles, transitioning to keypoint-based ironing once corners emerge. The keypoint model achieves a Mean Position Error (MPE) of 1.7615 pixels. The perception system transfers to physical fabrics without fine-tuning, outperforming baselines that fail in high-occlusion states or yield false positives on severe folds.

Xueyang Kang, Fengze Han, Abdur R. Fayjie et al.

Most existing methods for depth estimation from a focal stack of images employ convolutional neural networks (CNNs) using 2D or 3D convolutions over a fixed set of images. However, their effectiveness is constrained by the local properties of CNN kernels, which restricts them to process only focal stacks of fixed number of images during both training and inference. This limitation hampers their ability to generalize to stacks of arbitrary lengths. To overcome these limitations, we present a novel Transformer-based network, FocDepthFormer, which integrates a Transformer with an LSTM module and a CNN decoder. The Transformer's self-attention mechanism allows for the learning of more informative spatial features by implicitly performing non-local cross-referencing. The LSTM module is designed to integrate representations across image stacks of varying lengths. Additionally, we employ multi-scale convolutional kernels in an early-stage encoder to capture low-level features at different degrees of focus/defocus. By incorporating the LSTM, FocDepthFormer can be pre-trained on large-scale monocular RGB depth estimation datasets, improving visual pattern learning and reducing reliance on difficult-to-obtain focal stack data. Extensive experiments on diverse focal stack benchmark datasets demonstrate that our model outperforms state-of-the-art approaches across multiple evaluation metrics.

Xueyang Kang, Zhengkang Xiang, Zezheng Zhang et al.

Novel view synthesis (NVS) from a single image is highly ill-posed due to large unobserved regions, especially for views that deviate significantly from the input. While existing methods focus on consistency between the source and generated views, they often fail to maintain coherence and correct view alignment across long-range or looped trajectories. We propose a model that addresses this by decomposing single-view NVS into a 360-degree scene extrapolation followed by novel view interpolation. This design ensures long-term view and scene consistency by conditioning on keyframes extracted and warped from a generated panoramic representation. In the first stage, a panorama diffusion model learns the scene prior from the input perspective image. Perspective keyframes are then sampled and warped from the panorama and used as anchor frames in a pre-trained video diffusion model, which generates novel views through a proposed spatial noise diffusion process. Compared to prior work, our method produces globally consistent novel views -- even in loop closure scenarios -- while enabling flexible camera control. Experiments on diverse scene datasets demonstrate that our approach outperforms existing methods in generating coherent views along user-defined trajectories. Our implementation is available at https://github.com/YiGuYT/LookBeyond.

Xueyang Kang, Zijian Yu, Kourosh Khoshelham et al.

Interactive segmentation allows efficient label generation by leveraging user-provided clicks to progressively refine predictions, which is critical when fully supervised labels are costly or generalization to unseen classes is needed. Existing 3D interactive methods are limited: most operate sequentially, predicting only one object per iteration with binary masks, while several recent approaches depend on 2D foundation models and camera alignment to bridge the 2D-3D gap. To address these limitations, we propose a novel interactive segmentation framework that operates directly on sparse, randomly downsampled 3D points and processes multiple object clicks in a single forward pass. Our framework consists of a point Transformer-based encoder and a hierarchical mask decoder, which integrates multi-level crop-and-merge operations conditioned on learnable semantic embeddings. Unlike prior interactive approaches that require repeated model updates after each manually corrective click, our method jointly reasons over all click queries, modeling inter-instance relationships and refining both spatial masks and semantic predictions through spatial and semantic embeddings. Extensive experiments demonstrate that our model improves the mIoU metric by over 20 percent compared to strong baselines and achieves 8-10 percent gains under cross-dataset evaluation for a one-click per instance setting, often requiring only a single click per object. Our approach provides a generalizable and efficient solution for interactive 3D instance segmentation, particularly suitable for real-time applications such as robotic manipulation, navigation, and rapid 3D semantic annotation.

Lik Hang Kenny Wong, Xueyang Kang, Kaixin Bai et al.

Navigation and manipulation are core capabilities in Embodied AI, yet training agents with these capabilities in the real world faces high costs and time complexity. Therefore, sim-to-real transfer has emerged as a key approach, yet the sim-to-real gap persists. This survey examines how physics simulators address this gap by analyzing their properties overlooked in previous surveys. We also analyze their features for navigation and manipulation tasks, along with hardware requirements. Additionally, we offer a resource with benchmark datasets, metrics, simulation platforms, and cutting-edge methods-such as world models and geometric equivariance-to help researchers select suitable tools while accounting for hardware constraints.

Xueyang Kang, Zizhao Li, Tian Lan et al.

3D shape anomaly detection is a crucial task for industrial inspection and geometric analysis. Existing deep learning approaches typically learn representations of normal shapes and identify anomalies via out-of-distribution feature detection or decoder-based reconstruction. They often fail to generalize across diverse anomaly types and scales, such as global geometric errors (e.g., planar shifts, angle misalignments), and are sensitive to noisy or incomplete local points during training. To address these limitations, we propose a hierarchical point-patch anomaly scoring network that jointly models regional part features and local point features for robust anomaly reasoning. An adaptive patchification module integrates self-supervised decomposition to capture complex structural deviations. Beyond evaluations on public benchmarks (Anomaly-ShapeNet and Real3D-AD), we release an industrial test set with real CAD models exhibiting planar, angular, and structural defects. Experiments on public and industrial datasets show superior AUC-ROC and AUC-PR performance, including over 40% point-level improvement on the new industrial anomaly type and average object-level gains of 7% on Real3D-AD and 4% on Anomaly-ShapeNet, demonstrating strong robustness and generalization.

Zizhao Li, Xueyang Kang, Joseph West et al.

The ability to detect objects that are not prevalent in the training set is a critical capability in many 3D applications, including autonomous driving. Machine learning methods for object recognition often assume that all object categories encountered during inference belong to a closed set of classes present in the training data. This assumption limits generalization to the real world, as objects not seen during training may be misclassified or entirely ignored. As part of reliable AI, OOD detection identifies inputs that deviate significantly from the training distribution. This paper provides a comprehensive overview of OOD detection within the broader scope of trustworthy and uncertain AI. We begin with key use cases across diverse domains, introduce benchmark datasets spanning multiple modalities, and discuss evaluation metrics. Next, we present a comparative analysis of OOD detection methodologies, exploring model structures, uncertainty indicators, and distributional distance taxonomies, alongside uncertainty calibration techniques. Finally, we highlight promising research directions, including adversarially robust OOD detection and failure identification, particularly relevant to 3D applications. The paper offers both theoretical and practical insights into OOD detection, showcasing emerging research opportunities such as 3D vision integration. These insights help new researchers navigate the field more effectively, contributing to the development of reliable, safe, and robust AI systems.

Xueyang Kang

Modern deep learning developments create new opportunities for 3D mapping technology, scene reconstruction pipelines, and virtual reality development. Despite advances in 3D deep learning technology, direct training of deep learning models on 3D data faces challenges due to the high dimensionality inherent in 3D data and the scarcity of labeled datasets. Structure-from-motion (SfM) and Simultaneous Localization and Mapping (SLAM) exhibit robust performance when applied to structured indoor environments but often struggle with ambiguous features in unstructured environments. These techniques often struggle to generate detailed geometric representations effective for downstream tasks such as rendering and semantic analysis. Current limitations require the development of 3D representation methods that combine traditional geometric techniques with deep learning capabilities to generate robust geometry-aware deep learning models. The dissertation provides solutions to the fundamental challenges in 3D vision by developing geometric deep learning methods tailored for essential tasks such as camera pose estimation, point cloud registration, depth prediction, and 3D reconstruction. The integration of geometric priors or constraints, such as including depth information, surface normals, and equivariance into deep learning models, enhances both the accuracy and robustness of geometric representations. This study systematically investigates key components of 3D vision, including camera pose estimation, point cloud registration, depth estimation, and high-fidelity 3D reconstruction, demonstrating their effectiveness across real-world applications such as digital cultural heritage preservation and immersive VR/AR environments.

Xueyang Kang, Hang Zhao, Kourosh Khoshelham et al.

Point cloud registration is crucial for ensuring 3D alignment consistency of multiple local point clouds in 3D reconstruction for remote sensing or digital heritage. While various point cloud-based registration methods exist, both non-learning and learning-based, they ignore point orientations and point uncertainties, making the model susceptible to noisy input and aggressive rotations of the input point cloud like orthogonal transformation; thus, it necessitates extensive training point clouds with transformation augmentations. To address these issues, we propose a novel surfel-based pose learning regression approach. Our method can initialize surfels from Lidar point cloud using virtual perspective camera parameters, and learns explicit $\mathbf{SE(3)}$ equivariant features, including both position and rotation through $\mathbf{SE(3)}$ equivariant convolutional kernels to predict relative transformation between source and target scans. The model comprises an equivariant convolutional encoder, a cross-attention mechanism for similarity computation, a fully-connected decoder, and a non-linear Huber loss. Experimental results on indoor and outdoor datasets demonstrate our model superiority and robust performance on real point-cloud scans compared to state-of-the-art methods.

Yinglong Feng, Shuncheng Wu, Okan Köpüklü et al.

This paper studies unsupervised monocular depth prediction problem. Most of existing unsupervised depth prediction algorithms are developed for outdoor scenarios, while the depth prediction work in the indoor environment is still very scarce to our knowledge. Therefore, this work focuses on narrowing the gap by firstly evaluating existing approaches in the indoor environments and then improving the state-of-the-art design of architecture. Unlike typical outdoor training dataset, such as KITTI with motion constraints, data for indoor environment contains more arbitrary camera movement and short baseline between two consecutive images, which deteriorates the network training for the pose estimation. To address this issue, we propose two methods: Firstly, we propose a novel reconstruction loss function to constraint pose estimation, resulting in accuracy improvement of the predicted disparity map; secondly, we use an ensemble learning with a flipping strategy along with a median filter, directly taking operation on the output disparity map. We evaluate our approaches on the TUM RGB-D and self-collected datasets. The results have shown that both approaches outperform the previous state-of-the-art unsupervised learning approaches.

Xueyang Kang, Shunying Yuan

Simultaneous mapping and localization (SLAM) in an real indoor environment is still a challenging task. Traditional SLAM approaches rely heavily on low-level geometric constraints like corners or lines, which may lead to tracking failure in textureless surroundings or cluttered world with dynamic objects. In this paper, a compact semantic SLAM framework is proposed, with utilization of both geometric and object-level semantic constraints jointly, a more consistent mapping result, and more accurate pose estimation can be obtained. Two main contributions are presented int the paper, a) a robust and efficient SLAM data association and optimization framework is proposed, it models both discrete semantic labeling and continuous pose. b) a compact map representation, combining 2D Lidar map with object detection is presented. Experiments on public indoor datasets, TUM-RGBD, ICL-NUIM, and our own collected datasets prove the improving of SLAM robustness and accuracy compared to other popular SLAM systems, meanwhile a map maintenance efficiency can be achieved.

Xueyang Kang, Shengjiong Yin, Yinglong Fen

Lidar is extensively used in the industry and mass-market. Due to its measurement accuracy and insensitivity to illumination compared to cameras, It is applied onto a broad range of applications, like geodetic engineering, self driving cars or virtual reality. But the 3D Lidar with multi-beam is very expensive, and the massive measurements data can not be fully leveraged on some constrained platforms. The purpose of this paper is to explore the possibility of using cheap 2D Lidar off-the-shelf, to preform complex 3D Reconstruction, moreover, the generated 3D map quality is evaluated by our proposed metrics at the end. The 3D map is constructed in two ways, one way in which the scan is performed at known positions with an external rotary axis at another plane. The other way, in which the 2D Lidar for mapping and another 2D Lidar for localization are placed on a trolley, the trolley is pushed on the ground arbitrarily. The generated maps by different approaches are converted to octomaps uniformly before the evaluation. The similarity and difference between two maps will be evaluated by the proposed metrics thoroughly. The whole mapping system is composed of several modular components. A 3D bracket was made for assembling of the Lidar with a long range, the driver and the motor together. A cover platform made for the IMU and 2D Lidar with a shorter range but high accuracy. The software is stacked up in different ROS packages.