arXiv subjects

Xiaoshuai Zhang

Publications and source records attributed to Xiaoshuai Zhang.

At least 19 recordsLinked to original sources

arXiv2026

LVSPM: Long Sequence View Synthesis and Pose Estimation Model

We present LVSPM, a generalizable model that jointly estimates camera poses and synthesizes novel views from uncalibrated image collections. Trained with only RGB images and pose supervision, LVSPM avoids dense 3D ground truth and employs test-time training (TTT) layers to scale seamlessly to hundreds of input views. On RealEstate10k, Co3Dv2, and DL3DV, LVSPM surpasses VGGT in pose estimation across 16-256 views, with especially large margins at strict thresholds. For novel view synthesis under a practical protocol where more views cover larger scenes, LVSPM achieves state-of-the-art pose-free quality---surpassing even pose-dependent models in PSNR---and still maintains high quality as scene scale grows, while baselines collapse. The code is available at https://burningdust21.github.io/Projects/LVSPM .

cs.CV↗

arXiv2026

Physically-Guided Optical Inversion Enable Non-Contact Side-Channel Attack on Isolated Screens

Noncontact exfiltration of electronic screen content poses a security challenge, with side-channel incursions as the principal vector. We introduce an optical projection side-channel paradigm that confronts two core instabilities: (i) the near-singular Jacobian spectrum of projection mapping breaches Hadamard stability, rendering inversion hypersensitive to perturbations; (ii) irreversible compression in light transport obliterates global semantic cues, magnifying reconstruction ambiguity. Exploiting passive speckle patterns formed by diffuse reflection, our Irradiance Robust Radiometric Inversion Network (IR4Net) fuses a Physically Regularized Irradiance Approximation (PRIrr-Approximation), which embeds the radiative transfer equation in a learnable optimizer, with a contour-to-detail cross-scale reconstruction mechanism that arrests noise propagation. Moreover, an Irreversibility Constrained Semantic Reprojection (ICSR) module reinstates lost global structure through context-driven semantic mapping. Evaluated across four scene categories, IR4Net achieves fidelity beyond competing neural approaches while retaining resilience to illumination perturbations.

cs.CV↗

arXiv2026

PartUV: Part-Based UV Unwrapping of 3D Meshes

UV unwrapping flattens 3D surfaces to 2D with minimal distortion, often requiring the complex surface to be decomposed into multiple charts. Although extensively studied, existing UV unwrapping methods frequently struggle with AI-generated meshes, which are typically noisy, bumpy, and poorly conditioned. These methods often produce highly fragmented charts and suboptimal boundaries, introducing artifacts and hindering downstream tasks. We introduce PartUV, a part-based UV unwrapping pipeline that generates significantly fewer, part-aligned charts while maintaining low distortion. Built on top of a recent learning-based part decomposition method PartField, PartUV combines high-level semantic part decomposition with novel geometric heuristics in a top-down recursive framework. It ensures each chart's distortion remains below a user-specified threshold while minimizing the total number of charts. The pipeline integrates and extends parameterization and packing algorithms, incorporates dedicated handling of non-manifold and degenerate meshes, and is extensively parallelized for efficiency. Evaluated across four diverse datasets, including man-made, CAD, AI-generated, and Common Shapes, PartUV outperforms existing tools and recent neural methods in chart count and seam length, achieves comparable distortion, exhibits high success rates on challenging meshes, and enables new applications like part-specific multi-tiles packing. Our project page is at https://www.zhaoningwang.com/PartUV.

cs.CV↗

arXiv2025

Wavefront-Constrained Passive Obscured Object Detection

Accurately localizing and segmenting obscured objects from faint light patterns beyond the field of view is highly challenging due to multiple scattering and medium-induced perturbations. Most existing methods, based on real-valued modeling or local convolutional operations, are inadequate for capturing the underlying physics of coherent light propagation. Moreover, under low signal-to-noise conditions, these methods often converge to non-physical solutions, severely compromising the stability and reliability of the observation. To address these challenges, we propose a novel physics-driven Wavefront Propagating Compensation Network (WavePCNet) to simulate wavefront propagation and enhance the perception of obscured objects. This WavePCNet integrates the Tri-Phase Wavefront Complex-Propagation Reprojection (TriWCP) to incorporate complex amplitude transfer operators to precisely constrain coherent propagation behavior, along with a momentum memory mechanism to effectively suppress the accumulation of perturbations. Additionally, a High-frequency Cross-layer Compensation Enhancement is introduced to construct frequency-selective pathways with multi-scale receptive fields and dynamically model structural consistency across layers, further boosting the model's robustness and interpretability under complex environmental conditions. Extensive experiments conducted on four physically collected datasets demonstrate that WavePCNet consistently outperforms state-of-the-art methods across both accuracy and robustness.

cs.CV↗

arXiv2025

LARM: A Large Articulated-Object Reconstruction Model

Modeling 3D articulated objects with realistic geometry, textures, and kinematics is essential for a wide range of applications. However, existing optimization-based reconstruction methods often require dense multi-view inputs and expensive per-instance optimization, limiting their scalability. Recent feedforward approaches offer faster alternatives but frequently produce coarse geometry, lack texture reconstruction, and rely on brittle, complex multi-stage pipelines. We introduce LARM, a unified feedforward framework that reconstructs 3D articulated objects from sparse-view images by jointly recovering detailed geometry, realistic textures, and accurate joint structures. LARM extends LVSM a recent novel view synthesis (NVS) approach for static 3D objects into the articulated setting by jointly reasoning over camera pose and articulation variation using a transformer-based architecture, enabling scalable and accurate novel view synthesis. In addition, LARM generates auxiliary outputs such as depth maps and part masks to facilitate explicit 3D mesh extraction and joint estimation. Our pipeline eliminates the need for dense supervision and supports high-fidelity reconstruction across diverse object categories. Extensive experiments demonstrate that LARM outperforms state-of-the-art methods in both novel view and state synthesis as well as 3D articulated object reconstruction, generating high-quality meshes that closely adhere to the input images. project page: https://sylviayuan-sy.github.io/larm-site/

cs.CV↗

arXiv2024

A Mathematical Theory of Hyper-simplex Fractal Network for Blockchain: Part I

Blockchain technology holds promise for Web 3.0, but scalability remains a critical challenge. Here, we present a mathematical theory for a novel blockchain network topology based on fractal N-dimensional simplexes. This Hyper-simplex fractal network folds one-dimensional data blocks into geometric shapes, reflecting both underlying and overlaying network connectivities. Our approach offers near-infinite scalability, accommodating trillions of nodes while maintaining efficiency. We derive the mathematical foundations for generating and describing these network topologies, proving key properties such as node count, connectivity patterns, and fractal dimension. The resulting structure facilitates a hierarchical consensus mechanism and enables deterministic address mapping for rapid routing. This theoretical framework lays the groundwork for next-generation blockchain architectures, potentially revolutionizing large-scale decentralized systems. The Part I work was conducted between March and September 2024.

cs.NI↗

arXiv2024

ConDense: Consistent 2D/3D Pre-training for Dense and Sparse Features from Multi-View Images

To advance the state of the art in the creation of 3D foundation models, this paper introduces the ConDense framework for 3D pre-training utilizing existing pre-trained 2D networks and large-scale multi-view datasets. We propose a novel 2D-3D joint training scheme to extract co-embedded 2D and 3D features in an end-to-end pipeline, where 2D-3D feature consistency is enforced through a volume rendering NeRF-like ray marching process. Using dense per pixel features we are able to 1) directly distill the learned priors from 2D models to 3D models and create useful 3D backbones, 2) extract more consistent and less noisy 2D features, 3) formulate a consistent embedding space where 2D, 3D, and other modalities of data (e.g., natural language prompts) can be jointly queried. Furthermore, besides dense features, ConDense can be trained to extract sparse features (e.g., key points), also with 2D-3D consistency -- condensing 3D NeRF representations into compact sets of decorated key points. We demonstrate that our pre-trained model provides good initialization for various 3D tasks including 3D classification and segmentation, outperforming other 3D pre-training methods by a significant margin. It also enables, by exploiting our sparse features, additional useful downstream tasks, such as matching 2D images to 3D scenes, detecting duplicate 3D scenes, and querying a repository of 3D scenes through natural language -- all quite efficiently and without any per-scene fine-tuning.

cs.CV↗

arXiv2024

MeshFormer: High-Quality Mesh Generation with 3D-Guided Reconstruction Model

Open-world 3D reconstruction models have recently garnered significant attention. However, without sufficient 3D inductive bias, existing methods typically entail expensive training costs and struggle to extract high-quality 3D meshes. In this work, we introduce MeshFormer, a sparse-view reconstruction model that explicitly leverages 3D native structure, input guidance, and training supervision. Specifically, instead of using a triplane representation, we store features in 3D sparse voxels and combine transformers with 3D convolutions to leverage an explicit 3D structure and projective bias. In addition to sparse-view RGB input, we require the network to take input and generate corresponding normal maps. The input normal maps can be predicted by 2D diffusion models, significantly aiding in the guidance and refinement of the geometry's learning. Moreover, by combining Signed Distance Function (SDF) supervision with surface rendering, we directly learn to generate high-quality meshes without the need for complex multi-stage training processes. By incorporating these explicit 3D biases, MeshFormer can be trained efficiently and deliver high-quality textured meshes with fine-grained geometric details. It can also be integrated with 2D diffusion models to enable fast single-image-to-3D and text-to-3D tasks. Project page: https://meshformer3d.github.io

cs.CV↗

arXiv2024

An Origami-Inspired Endoscopic Capsule with Tactile Perception for Early Tissue Anomaly Detection

Video Capsule Endoscopy (VCE) is currently one of the most effective methods for detecting intestinal diseases. However, it is challenging to detect early-stage small nodules with this method because they lack obvious color or shape features. In this letter, we present a new origami capsule endoscope to detect early small intestinal nodules using tactile sensing. Four soft tactile sensors made out of piezoresistive material feed four channels of phase-shifted data that are processed using a particle filter. The particle filter uses an importance assignment template designed using experimental data from five known sizes of modules. Moreover, the proposed capsule can use shape changes to passively move forward or backward under peristalsis, enabling it to reach any position in the intestine for detection. Experimental results show that the proposed capsule can detect nodules of more than 3mm diameter with 100% accuracy.

eess.SY↗

arXiv2024

TensoIR: Tensorial Inverse Rendering

We propose TensoIR, a novel inverse rendering approach based on tensor factorization and neural fields. Unlike previous works that use purely MLP-based neural fields, thus suffering from low capacity and high computation costs, we extend TensoRF, a state-of-the-art approach for radiance field modeling, to estimate scene geometry, surface reflectance, and environment illumination from multi-view images captured under unknown lighting conditions. Our approach jointly achieves radiance field reconstruction and physically-based model estimation, leading to photo-realistic novel view synthesis and relighting results. Benefiting from the efficiency and extensibility of the TensoRF-based representation, our method can accurately model secondary shading effects (like shadows and indirect lighting) and generally support input images captured under single or multiple unknown lighting conditions. The low-rank tensor representation allows us to not only achieve fast and compact reconstruction but also better exploit shared information under an arbitrary number of capturing lighting conditions. We demonstrate the superiority of our method to baseline methods qualitatively and quantitatively on various challenging synthetic and real-world scenes.

cs.CV↗

arXiv2023

When Web 3.0 Meets Reality: A Hyperdimensional Fractal Polytope P2P Ecosystems

Web 3.0 opens the world of new existence of the crypto-network-entity, which is independently defined by the public key pairs for entities and the connection to the Web 3.0 cyberspace. In this paper, we first discover a spacetime coordinate system based on fractal polytope in any dimensions with discrete time offered by blockchain and consensus. Second, the novel network entities and functions are defined to make use of hyperdimensional deterministic switching and routing protocols and blockchain-enabled mutual authentication. In addition to spacetime network architecture, we also define a multi-tier identity scheme which extends the native Web 3.0 crypto-network-entity to outer cyber and physical world, offering legal-compliant anonymity and linkability to all derived identifiers of entities. In this way, we unify the holistic Web 3.0 network based on persistent spacetime and its entity extension to our cyber and physical world.

cs.NI↗

arXiv2023

A Review of Gaps between Web 4.0 and Web 3.0 Intelligent Network Infrastructure

World Wide Web is speeding up its pace into an intelligent and decentralized ecosystem, as seen in the campaign of Web 3.0 and forthcoming Web 4.0. Marked by the Europe Commission's latest mention of Web 4.0, a race towards strategic Web 4.0 success has started. Web 4.0 is committed to bringing the next technological transition with an open, secure, trustworthy fairness and digital ecosystem for individuals and businesses in private and public sectors. Despite overlapping scopes and objectives of Web 3.0 and Web 4.0 from academic and industrial perspectives, there are distinct and definitive features and gaps for the next generation of WWW. In this review, a brief introduction to WWW development unravels the entangled but consistent requirement of a more vivid web experience, enhancing human-centric experience in both societal and technical aspects. Moreover, the review brings a decentralized intelligence prospect of view on native AI entities for Web 4.0, envisioning sustainable, autonomous and decentralized AI services for the entire Web 4.0 environment, powering a self-sustainable Decentralized Physical and Software Infrastructure for Computing Force Network, Semantic Network, Virtual/Mixed Reality, and Privacy-preserving content presumption. The review aims to reveal that Web 4.0 offers native intelligence with focused thinking on utilizing decentralized physical infrastructure, in addition to sole requirements on decentralization, bridging the gap between Web 4.0 and Web 3.0 advances with the latest future-shaping blockchain-enabled computing and network routing protocols.

cs.NI↗

arXiv2023

deController: A Web3 Native Cyberspace Infrastructure Perspective

Web3 brings an emerging outlook for the value of decentralization, boosting the decentralized infrastructure. People can benefit from Web3, facilitated by the advances in distributed ledger technology, to read, write and own web content, services and applications more freely without revealing their real identities. Although the features and merits of Web3 have been widely discussed, the network architecture of Web3 and how to achieve complete decentralization considering law compliance in Web3 are still unclear. Here, we propose a perspective of Web3 architecture, deController, consisting of underlay and overlay network as Web3 infrastructures to underpin services and applications. The functions of underlay and overlay and their interactions are illustrated. Meanwhile, the security and privacy of Web3 are analyzed based on a novel design of three-tier identities cooperating with deController. Furthermore, the impacts of laws on privacy and cyber sovereignty to achieve Web3 are discussed.

cs.NI↗

arXiv2023

MovingParts: Motion-based 3D Part Discovery in Dynamic Radiance Field

We present MovingParts, a NeRF-based method for dynamic scene reconstruction and part discovery. We consider motion as an important cue for identifying parts, that all particles on the same part share the common motion pattern. From the perspective of fluid simulation, existing deformation-based methods for dynamic NeRF can be seen as parameterizing the scene motion under the Eulerian view, i.e., focusing on specific locations in space through which the fluid flows as time passes. However, it is intractable to extract the motion of constituting objects or parts using the Eulerian view representation. In this work, we introduce the dual Lagrangian view and enforce representations under the Eulerian/Lagrangian views to be cycle-consistent. Under the Lagrangian view, we parameterize the scene motion by tracking the trajectory of particles on objects. The Lagrangian view makes it convenient to discover parts by factorizing the scene motion as a composition of part-level rigid motions. Experimentally, our method can achieve fast and high-quality dynamic scene reconstruction from even a single moving camera, and the induced part-based representation allows direct applications of part tracking, animation, 3D scene editing, etc.

cs.CV↗

arXiv2023

Nerflets: Local Radiance Fields for Efficient Structure-Aware 3D Scene Representation from 2D Supervision

We address efficient and structure-aware 3D scene representation from images. Nerflets are our key contribution -- a set of local neural radiance fields that together represent a scene. Each nerflet maintains its own spatial position, orientation, and extent, within which it contributes to panoptic, density, and radiance reconstructions. By leveraging only photometric and inferred panoptic image supervision, we can directly and jointly optimize the parameters of a set of nerflets so as to form a decomposed representation of the scene, where each object instance is represented by a group of nerflets. During experiments with indoor and outdoor environments, we find that nerflets: (1) fit and approximate the scene more efficiently than traditional global NeRFs, (2) allow the extraction of panoptic and photometric renderings from arbitrary views, and (3) enable tasks rare for NeRFs, such as 3D panoptic segmentation and interactive editing.

cs.CV↗

arXiv2023

Close the Optical Sensing Domain Gap by Physics-Grounded Active Stereo Sensor Simulation

In this paper, we focus on the simulation of active stereovision depth sensors, which are popular in both academic and industry communities. Inspired by the underlying mechanism of the sensors, we designed a fully physics-grounded simulation pipeline that includes material acquisition, ray-tracing-based infrared (IR) image rendering, IR noise simulation, and depth estimation. The pipeline is able to generate depth maps with material-dependent error patterns similar to a real depth sensor in real time. We conduct real experiments to show that perception algorithms and reinforcement learning policies trained in our simulation platform could transfer well to the real-world test cases without any fine-tuning. Furthermore, due to the high degree of realism of this simulation, our depth sensor simulator can be used as a convenient testbed to evaluate the algorithm performance in the real world, which will largely reduce the human effort in developing robotic algorithms. The entire pipeline has been integrated into the SAPIEN simulator and is open-sourced to promote the research of vision and robotics communities.

cs.RO↗

arXiv2022

Style Equalization: Unsupervised Learning of Controllable Generative Sequence Models

Controllable generative sequence models with the capability to extract and replicate the style of specific examples enable many applications, including narrating audiobooks in different voices, auto-completing and auto-correcting written handwriting, and generating missing training samples for downstream recognition tasks. However, under an unsupervised-style setting, typical training algorithms for controllable sequence generative models suffer from the training-inference mismatch, where the same sample is used as content and style input during training but unpaired samples are given during inference. In this paper, we tackle the training-inference mismatch encountered during unsupervised learning of controllable generative sequence models. The proposed method is simple yet effective, where we use a style transformation module to transfer target style information into an unrelated style input. This method enables training using unpaired content and style samples and thereby mitigate the training-inference mismatch. We apply style equalization to text-to-speech and text-to-handwriting synthesis on three datasets. We conduct thorough evaluation, including both quantitative and qualitative user studies. Our results show that by mitigating the training-inference mismatch with the proposed style equalization, we achieve style replication scores comparable to real data in our user studies.

cs.LG↗

arXiv2022

NeRFusion: Fusing Radiance Fields for Large-Scale Scene Reconstruction

While NeRF has shown great success for neural reconstruction and rendering, its limited MLP capacity and long per-scene optimization times make it challenging to model large-scale indoor scenes. In contrast, classical 3D reconstruction methods can handle large-scale scenes but do not produce realistic renderings. We propose NeRFusion, a method that combines the advantages of NeRF and TSDF-based fusion techniques to achieve efficient large-scale reconstruction and photo-realistic rendering. We process the input image sequence to predict per-frame local radiance fields via direct network inference. These are then fused using a novel recurrent neural network that incrementally reconstructs a global, sparse scene representation in real-time at 22 fps. This global volume can be further fine-tuned to boost rendering quality. We demonstrate that NeRFusion achieves state-of-the-art quality on both large-scale indoor and small-scale object scenes, with substantially faster reconstruction than NeRF and other recent methods.

cs.CV↗