arXiv subjects

Zhiyuan Zhang

Publications and source records attributed to Zhiyuan Zhang.

At least 91 records · Page 5Linked to original sources

arXiv2025

HDiffTG: A Lightweight Hybrid Diffusion-Transformer-GCN Architecture for 3D Human Pose Estimation

We propose HDiffTG, a novel 3D Human Pose Estimation (3DHPE) method that integrates Transformer, Graph Convolutional Network (GCN), and diffusion model into a unified framework. HDiffTG leverages the strengths of these techniques to significantly improve pose estimation accuracy and robustness while maintaining a lightweight design. The Transformer captures global spatiotemporal dependencies, the GCN models local skeletal structures, and the diffusion model provides step-by-step optimization for fine-tuning, achieving a complementary balance between global and local features. This integration enhances the model's ability to handle pose estimation under occlusions and in complex scenarios. Furthermore, we introduce lightweight optimizations to the integrated model and refine the objective function design to reduce computational overhead without compromising performance. Evaluation results on the Human3.6M and MPI-INF-3DHP datasets demonstrate that HDiffTG achieves state-of-the-art (SOTA) performance on the MPI-INF-3DHP dataset while excelling in both accuracy and computational efficiency. Additionally, the model exhibits exceptional robustness in noisy and occluded environments. Source codes and models are available at https://github.com/CirceJie/HDiffTG

cs.CV↗

arXiv2025

TS-Diff: Two-Stage Diffusion Model for Low-Light RAW Image Enhancement

This paper presents a novel Two-Stage Diffusion Model (TS-Diff) for enhancing extremely low-light RAW images. In the pre-training stage, TS-Diff synthesizes noisy images by constructing multiple virtual cameras based on a noise space. Camera Feature Integration (CFI) modules are then designed to enable the model to learn generalizable features across diverse virtual cameras. During the aligning stage, CFIs are averaged to create a target-specific CFI$^T$, which is fine-tuned using a small amount of real RAW data to adapt to the noise characteristics of specific cameras. A structural reparameterization technique further simplifies CFI$^T$ for efficient deployment. To address color shifts during the diffusion process, a color corrector is introduced to ensure color consistency by dynamically adjusting global color distributions. Additionally, a novel dataset, QID, is constructed, featuring quantifiable illumination levels and a wide dynamic range, providing a comprehensive benchmark for training and evaluation under extreme low-light conditions. Experimental results demonstrate that TS-Diff achieves state-of-the-art performance on multiple datasets, including QID, SID, and ELD, excelling in denoising, generalization, and color consistency across various cameras and illumination levels. These findings highlight the robustness and versatility of TS-Diff, making it a practical solution for low-light imaging applications. Source codes and models are available at https://github.com/CircccleK/TS-Diff

cs.CV↗

arXiv2025

RoboAct-CLIP: Video-Driven Pre-training of Atomic Action Understanding for Robotics

Visual Language Models (VLMs) have emerged as pivotal tools for robotic systems, enabling cross-task generalization, dynamic environmental interaction, and long-horizon planning through multimodal perception and semantic reasoning. However, existing open-source VLMs predominantly trained for generic vision-language alignment tasks fail to model temporally correlated action semantics that are crucial for robotic manipulation effectively. While current image-based fine-tuning methods partially adapt VLMs to robotic applications, they fundamentally disregard temporal evolution patterns in video sequences and suffer from visual feature entanglement between robotic agents, manipulated objects, and environmental contexts, thereby limiting semantic decoupling capability for atomic actions and compromising model generalizability.To overcome these challenges, this work presents RoboAct-CLIP with dual technical contributions: 1) A dataset reconstruction framework that performs semantic-constrained action unit segmentation and re-annotation on open-source robotic videos, constructing purified training sets containing singular atomic actions (e.g., "grasp"); 2) A temporal-decoupling fine-tuning strategy based on Contrastive Language-Image Pretraining (CLIP) architecture, which disentangles temporal action features across video frames from object-centric characteristics to achieve hierarchical representation learning of robotic atomic actions.Experimental results in simulated environments demonstrate that the RoboAct-CLIP pretrained model achieves a 12% higher success rate than baseline VLMs, along with superior generalization in multi-object manipulation tasks.

cs.RO↗

arXiv2025

MPDrive: Improving Spatial Understanding with Marker-Based Prompt Learning for Autonomous Driving

Autonomous driving visual question answering (AD-VQA) aims to answer questions related to perception, prediction, and planning based on given driving scene images, heavily relying on the model's spatial understanding capabilities. Prior works typically express spatial information through textual representations of coordinates, resulting in semantic gaps between visual coordinate representations and textual descriptions. This oversight hinders the accurate transmission of spatial information and increases the expressive burden. To address this, we propose a novel Marker-based Prompt learning framework (MPDrive), which represents spatial coordinates by concise visual markers, ensuring linguistic expressive consistency and enhancing the accuracy of both visual perception and spatial expression in AD-VQA. Specifically, we create marker images by employing a detection expert to overlay object regions with numerical labels, converting complex textual coordinate generation into straightforward text-based visual marker predictions. Moreover, we fuse original and marker images as scene-level features and integrate them with detection priors to derive instance-level features. By combining these features, we construct dual-granularity visual prompts that stimulate the LLM's spatial perception capabilities. Extensive experiments on the DriveLM and CODA-LM datasets show that MPDrive achieves state-of-the-art performance, particularly in cases requiring sophisticated spatial understanding.

cs.CV↗

arXiv2025

Learning Orientation Field for OSM-Guided Autonomous Navigation

OpenStreetMap (OSM) has gained popularity recently in autonomous navigation due to its public accessibility, lower maintenance costs, and broader geographical coverage. However, existing methods often struggle with noisy OSM data and incomplete sensor observations, leading to inaccuracies in trajectory planning. These challenges are particularly evident in complex driving scenarios, such as at intersections or facing occlusions. To address these challenges, we propose a robust and explainable two-stage framework to learn an Orientation Field (OrField) for robot navigation by integrating LiDAR scans and OSM routes. In the first stage, we introduce the novel representation, OrField, which can provide orientations for each grid on the map, reasoning jointly from noisy LiDAR scans and OSM routes. To generate a robust OrField, we train a deep neural network by encoding a versatile initial OrField and output an optimized OrField. Based on OrField, we propose two trajectory planners for OSM-guided robot navigation, called Field-RRT* and Field-Bezier, respectively, in the second stage by improving the Rapidly Exploring Random Tree (RRT) algorithm and Bezier curve to estimate the trajectories. Thanks to the robustness of OrField which captures both global and local information, Field-RRT* and Field-Bezier can generate accurate and reliable trajectories even in challenging conditions. We validate our approach through experiments on the SemanticKITTI dataset and our own campus dataset. The results demonstrate the effectiveness of our method, achieving superior performance in complex and noisy conditions. Our code for network training and real-world deployment is available at https://github.com/IMRL/OriField.

cs.RO↗

arXiv2025

Judge Anything: MLLM as a Judge Across Any Modality

Evaluating generative foundation models on open-ended multimodal understanding (MMU) and generation (MMG) tasks across diverse modalities (e.g., images, audio, video) poses significant challenges due to the complexity of cross-modal interactions. To this end, the idea of utilizing Multimodal LLMs (MLLMs) as automated judges has emerged, with encouraging results in assessing vision-language understanding tasks. Moving further, this paper extends MLLM-as-a-Judge across modalities to a unified manner by introducing two benchmarks, TaskAnything and JudgeAnything, to respectively evaluate the overall performance and judging capabilities of MLLMs across any-to-any modality tasks. Specifically, TaskAnything evaluates the MMU and MMG capabilities across 15 any-to-any modality categories, employing 1,500 queries curated from well-established benchmarks. Furthermore, JudgeAnything evaluates the judging capabilities of 5 advanced (e.g., GPT-4o and Gemini-2.0-Flash) from the perspectives of Pair Comparison and Score Evaluation, providing a standardized testbed that incorporates human judgments and detailed rubrics. Our extensive experiments reveal that while these MLLMs show promise in assessing MMU (i.e., achieving an average of 66.55% in Pair Comparison setting and 42.79% in Score Evaluation setting), they encounter significant challenges with MMG tasks (i.e., averaging only 53.37% in Pair Comparison setting and 30.05% in Score Evaluation setting), exposing cross-modality biases and hallucination issues. To address this, we present OmniArena, an automated platform for evaluating omni-models and multimodal reward models. Our work highlights the need for fairer evaluation protocols and stronger alignment with human preferences. The source code and dataset are publicly available at: https://urrealhero.github.io/judgeanythingweb/.

cs.CL↗

arXiv2025

I2V3D: Controllable image-to-video generation with 3D guidance

We present I2V3D, a novel framework for animating static images into dynamic videos with precise 3D control, leveraging the strengths of both 3D geometry guidance and advanced generative models. Our approach combines the precision of a computer graphics pipeline, enabling accurate control over elements such as camera movement, object rotation, and character animation, with the visual fidelity of generative AI to produce high-quality videos from coarsely rendered inputs. To support animations with any initial start point and extended sequences, we adopt a two-stage generation process guided by 3D geometry: 1) 3D-Guided Keyframe Generation, where a customized image diffusion model refines rendered keyframes to ensure consistency and quality, and 2) 3D-Guided Video Interpolation, a training-free approach that generates smooth, high-quality video frames between keyframes using bidirectional guidance. Experimental results highlight the effectiveness of our framework in producing controllable, high-quality animations from single input images by harmonizing 3D geometry with generative models. The code for our framework will be publicly released.

cs.CV↗

arXiv2025

Closed-Loop Open-Vocabulary Mobile Manipulation with GPT-4V

Autonomous robot navigation and manipulation in open environments require reasoning and replanning with closed-loop feedback. In this work, we present COME-robot, the first closed-loop robotic system utilizing the GPT-4V vision-language foundation model for open-ended reasoning and adaptive planning in real-world scenarios.COME-robot incorporates two key innovative modules: (i) a multi-level open-vocabulary perception and situated reasoning module that enables effective exploration of the 3D environment and target object identification using commonsense knowledge and situated information, and (ii) an iterative closed-loop feedback and restoration mechanism that verifies task feasibility, monitors execution success, and traces failure causes across different modules for robust failure recovery. Through comprehensive experiments involving 8 challenging real-world mobile and tabletop manipulation tasks, COME-robot demonstrates a significant improvement in task success rate (~35%) compared to state-of-the-art methods. We further conduct comprehensive analyses to elucidate how COME-robot's design facilitates failure recovery, free-form instruction following, and long-horizon task planning.

cs.RO↗

arXiv2025

BHViT: Binarized Hybrid Vision Transformer

Model binarization has made significant progress in enabling real-time and energy-efficient computation for convolutional neural networks (CNN), offering a potential solution to the deployment challenges faced by Vision Transformers (ViTs) on edge devices. However, due to the structural differences between CNN and Transformer architectures, simply applying binary CNN strategies to the ViT models will lead to a significant performance drop. To tackle this challenge, we propose BHViT, a binarization-friendly hybrid ViT architecture and its full binarization model with the guidance of three important observations. Initially, BHViT utilizes the local information interaction and hierarchical feature aggregation technique from coarse to fine levels to address redundant computations stemming from excessive tokens. Then, a novel module based on shift operations is proposed to enhance the performance of the binary Multilayer Perceptron (MLP) module without significantly increasing computational overhead. In addition, an innovative attention matrix binarization method based on quantization decomposition is proposed to evaluate the token's importance in the binarized attention matrix. Finally, we propose a regularization loss to address the inadequate optimization caused by the incompatibility between the weight oscillation in the binary layers and the Adam Optimizer. Extensive experimental results demonstrate that our proposed algorithm achieves SOTA performance among binary ViT methods.

cs.CV↗

arXiv2025

A mixed precision preconditioned Jacobi method for the symmetric eigenvalue problem

The eigenvalue problem is a fundamental problem in scientific computing. In this paper, we first give the error analysis for a single step or sweep of Jacobi's method in floating point arithmetic. Then we propose a mixed precision preconditioned Jacobi method for the symmetric eigenvalue problem: We first compute the eigenvalue decomposition of a real symmetric matrix by an eigensolver at low precision and we obtain a low-precision matrix of eigenvectors; Then by using the high-precision modified Gram-Schmidt orthogonalization process, a high-precision orthogonal matrix is obtained, which is used as an initial guess for Jacobi's method. The rounding error analysis of the proposed method is established under some conditions. We also present a mixed precision preconditioned one-sided Jacobi method for the singular value problem and the corresponding rounding error analysis is discussed. Numerical experiments on CPUs and GPUs are reported to illustrate the efficiency of the proposed method over the original Jacobi method.

math.NA↗

arXiv2025

Evidence of P-wave Pairing in K$_2$Cr$_3$As$_3$ Superconductors from Phase-sensitive Measurement

P-wave superconductors hold immense promise for both fundamental physics and practical applications due to their unusual pairing symmetry and potential topological superconductivity. However, the exploration of the p-wave superconductors has proved to be a complex endeavor. Not only are they rare in nature but also the identification of p-wave superconductors has been an arduous task in history. For example, phase-sensitive measurement, an experimental technique which can provide conclusive evidence for unconventional pairing, has not been implemented successfully to identify p-wave superconductors. Here, we study a recently discovered family of superconductors, A$_2$Cr$_3$As$_3$ (A = K, Rb, Cs), which were proposed theoretically to be a candidate of p-wave superconductors. We fabricate superconducting quantum interference devices (SQUIDs) on exfoliated K$_2$Cr$_3$As$_3$, and perform the phase-sensitive measurement. We observe that such SQUIDs exhibit a pronounced second-order harmonic component sin(2$π$) in the current-phase relation, suggesting the admixture of 0- and $π$-phase. By carefully examining the magnetic field dependence of the oscillation patterns of critical current and Shapiro steps under microwave irradiation, we reveal a crossover from 0- to $π$-dominating phase state and conclude that the existence of the $π$-phase is in favor of the p-wave pairing symmetry in K$_2$Cr$_3$As$_3$.

cond-mat.supr-con↗

arXiv2025

Observation of topological Anderson Chern insulator phase in MnBi$_4$Te$_7$ monolayer

The correlation of topology and disorder has attracted great intention due to appropriate disorder could induce the phase transition between trivial and nontrivial topological states. While it is widely recognized that strong disorder can produce rich phase diagrams in topological nontrivial states, moderate disorder has been proposed to induce transitions into topologically nontrivial phases counter-intuitively, leading to the concept of topological Anderson insulators. This phenomenon has been theoretically explored and simulated in various systems, yet experimental realization in solid state systems has remained elusive due to challenges in controlling disorder. Here, we report the experimental observation of Chern insulator state signed by the coexistence of quantized Hall plateau and zero longitudinal resistance in monolayer MnBi$_4$Te$_7$ Hall bar device, which originally hosts a trivial insulating state with Chern number $C$ = 0 in clean limit. We demonstrate that the observed trivial to nontrivial transition in this monolayer device can be attributed to disorder, evidenced by universal conductance fluctuations. Our findings substantiate the existence of a long-sought topological Anderson Chern insulator in real materials, a unique variant of the topological Anderson insulator characterized by broken time-reversal-symmetry.

cond-mat.mes-hall↗

arXiv2025

Implicit-Explicit simulation of Mass-Spring-Charge Systems

Point masses connected by springs, or mass-spring systems, are widely used in computer animation to approximate the behavior of deformable objects. One of the restrictions imposed by these models is that points that are not topologically constrained (linked by a spring) are unable to interact with each other explicitly. Such interactions would introduce a new dimension for artistic control and animation within the computer graphics community. Beyond graphics, such a model could be an effective proxy to use for model-based learning of complex physical systems such as molecular biology. We propose to imbue masses in a mass-spring system with electrostatic charge leading a system with internal forces between all pairs of charged points -- regardless of whether they are linked by a spring. We provide a practical and stable algorithm to simulate charged mass-spring systems over long time horizons. We demonstrate how these systems may be controlled via parameters such as guidance electric fields or external charges, thus presenting fresh opportunities for artistic authoring. Our method is especially appropriate for computer graphics applications due to its robustness at larger simulation time steps.

cs.GR↗

arXiv2024

Bench2Drive-R: Turning Real World Data into Reactive Closed-Loop Autonomous Driving Benchmark by Generative Model

For end-to-end autonomous driving (E2E-AD), the evaluation system remains an open problem. Existing closed-loop evaluation protocols usually rely on simulators like CARLA being less realistic; while NAVSIM using real-world vision data, yet is limited to fixed planning trajectories in short horizon and assumes other agents are not reactive. We introduce Bench2Drive-R, a generative framework that enables reactive closed-loop evaluation. Unlike existing video generative models for AD, the proposed designs are tailored for interactive simulation, where sensor rendering and behavior rollout are decoupled by applying a separate behavioral controller to simulate the reactions of surrounding agents. As a result, the renderer could focus on image fidelity, control adherence, and spatial-temporal coherence. For temporal consistency, due to the step-wise interaction nature of simulation, we design a noise modulating temporal encoder with Gaussian blurring to encourage long-horizon autoregressive rollout of image sequences without deteriorating distribution shifts. For spatial consistency, a retrieval mechanism, which takes the spatially nearest images as references, is introduced to to ensure scene-level rendering fidelity during the generation process. The spatial relations between target and reference are explicitly modeled with 3D relative position encodings and the potential over-reliance of reference images is mitigated with hierarchical sampling and classifier-free guidance. We compare the generation quality of Bench2Drive-R with existing generative models and achieve state-of-the-art performance. We further integrate Bench2Drive-R into nuPlan and evaluate the generative qualities with closed-loop simulation results. We will open source our code.

cs.RO↗

arXiv2024

Hypergraph burning, matchings, and zero forcing

Lazy burning is a recently introduced variation of burning where only one set of vertices is chosen to burn in the first round. In hypergraphs, lazy burning spreads when all but one vertex in a hyperedge is burned. The lazy burning number is the minimum number of initially burned vertices that eventually burns all vertices. We give several equivalent characterizations of lazy burning on hypergraphs using matchings and zero forcing, and then apply these to give new bounds and complexity results. We prove that the lazy burning number of a hypergraph $H$ equals its order minus the maximum cardinality of a certain matching on its incidence graph. Using this characterization, we give a formula for the lazy burning number of a dual hypergraph and give new bounds on the lazy burning number based on various hypergraph parameters. We show that the lazy burning number of a hypergraph may be characterized by a maximal subhypergraph that results from iteratively deleting vertices in singleton hyperedges. We prove that lazy burning on a hypergraph is equivalent to zero forcing on its incidence graph and show an equivalence between skew zero forcing on a graph and lazy burning on its neighborhood hypergraph. As a result, we show that finding an upper bound on the lazy burning number of a hypergraph is NP-complete, which resolves a conjecture from \cite{BJR}. By applying lazy burning, we show that computing an upper bound on the skew zero forcing number for bipartite graphs is NP-complete. We finish with open problems.

math.CO↗

arXiv2024

Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving

In an era marked by the rapid scaling of foundation models, autonomous driving technologies are approaching a transformative threshold where end-to-end autonomous driving (E2E-AD) emerges due to its potential of scaling up in the data-driven manner. However, existing E2E-AD methods are mostly evaluated under the open-loop log-replay manner with L2 errors and collision rate as metrics (e.g., in nuScenes), which could not fully reflect the driving performance of algorithms as recently acknowledged in the community. For those E2E-AD methods evaluated under the closed-loop protocol, they are tested in fixed routes (e.g., Town05Long and Longest6 in CARLA) with the driving score as metrics, which is known for high variance due to the unsmoothed metric function and large randomness in the long route. Besides, these methods usually collect their own data for training, which makes algorithm-level fair comparison infeasible. To fulfill the paramount need of comprehensive, realistic, and fair testing environments for Full Self-Driving (FSD), we present Bench2Drive, the first benchmark for evaluating E2E-AD systems' multiple abilities in a closed-loop manner. Bench2Drive's official training data consists of 2 million fully annotated frames, collected from 13638 short clips uniformly distributed under 44 interactive scenarios (cut-in, overtaking, detour, etc), 23 weathers (sunny, foggy, rainy, etc), and 12 towns (urban, village, university, etc) in CARLA v2. Its evaluation protocol requires E2E-AD models to pass 44 interactive scenarios under different locations and weathers which sums up to 220 routes and thus provides a comprehensive and disentangled assessment about their driving capability under different situations. We implement state-of-the-art E2E-AD models and evaluate them in Bench2Drive, providing insights regarding current status and future directions.

cs.RO↗

arXiv2024

Zimmer's conjecture for non-split semisimple Lie groups

We prove many new cases of Zimmer's conjecture for actions by lattices in non-$\mathbb{R}$-split semisimple Lie groups $G$. By prior arguments, Zimmer's conjecture reduces to studying certain probability measures invariant under a minimal parabolic subgroup for the induced $G$-action. Two techniques are introduced to give lower bounds on the dimension of a manifold $M$ admitting a non-isometric action. First, when the Levi component of the stabilizer of the measure has higher-rank simple factors, cocycle superrigidity provides a lower bound on the dimension of $M$. Second, when certain fiberwise coarse Lyapunov distributions are one-dimensional, a measure rigidity argument provides additional invariance of the measure if the associated root spaces are higher-dimensional.

math.DS↗

arXiv2024

Geometric properties of partially hyperbolic measures and applications to measure rigidity

We give a geometric characterization of the quantitative non-integrability, introduced by Katz, of strong stable and unstable bundles of partially hyperbolic measures and sets in dimension 3. This is done via the use of higher order templates for the invariant bundles. Using the recent work of Katz, we derive some consequences, including the measure rigidity of $uu$-states and the existence of physical measures.

math.DS↗