arXiv subjects

Zhe Li

Publications and source records attributed to Zhe Li.

At least 181 records · Page 10Linked to original sources

arXiv2024

MCMat: Multiview-Consistent and Physically Accurate PBR Material Generation

Existing 2D methods utilize UNet-based diffusion models to generate multi-view physically-based rendering (PBR) maps but struggle with multi-view inconsistency, while some 3D methods directly generate UV maps, encountering generalization issues due to the limited 3D data. To address these problems, we propose a two-stage approach, including multi-view generation and UV materials refinement. In the generation stage, we adopt a Diffusion Transformer (DiT) model to generate PBR materials, where both the specially designed multi-branch DiT and reference-based DiT blocks adopt a global attention mechanism to promote feature interaction and fusion between different views, thereby improving multi-view consistency. In addition, we adopt a PBR-based diffusion loss to ensure that the generated materials align with realistic physical principles. In the refinement stage, we propose a material-refined DiT that performs inpainting in empty areas and enhances details in UV space. Except for the normal condition, this refinement also takes the material map from the generation stage as an additional condition to reduce the learning difficulty and improve generalization. Extensive experiments show that our method achieves state-of-the-art performance in texturing 3D objects with PBR materials and provides significant advantages for graphics relighting applications. Project Page: https://lingtengqiu.github.io/2024/MCMat/

cs.CV↗

arXiv2024

Movable Antenna Aided NOMA: Joint Antenna Positioning, Precoding, and Decoding Design

This paper investigates movable antenna (MA) aided non-orthogonal multiple access (NOMA) for multi-user downlink communication, where the base station (BS) is equipped with a fixed-position antenna (FPA) array to serve multiple MA-enabled users. An optimization problem is formulated to maximize the minimum achievable rate among all the users by jointly optimizing the MA positioning of each user, the precoding matrix at the BS, and the successive interference cancellation (SIC) decoding indicator matrix at the users, subject to a set of constraints including the limited movement area of the MAs, the maximum transmit power of the BS, and the SIC decoding condition. To solve this non-convex problem, we propose a two-loop iterative optimization algorithm that combines the hippopotamus optimization (HO) method with the alternating optimization (AO) method to obtain a suboptimal solution efficiently. Specifically, in the inner loop, the complex-valued precoding matrix and the binary decoding indicator matrix are optimized alternatively by the successive convex approximation (SCA) technique with customized greedy search to maximize the minimum achievable rate for the given positions of the MAs. In the outer loop, each user's antenna position is updated using the HO algorithm, following a novel nature-inspired intelligent optimization framework. Simulation results show that the proposed algorithms can effectively avoid local optimum for highly coupled variables and significantly improve the rate performance of the NOMA system compared to the conventional FPA system as well as other benchmark schemes.

cs.IT↗

arXiv2024

Mixed QCD-EW corrections to $W$-pair production at electron-positron colliders

The discrepancy between the CDF measurement and the Standard Model theoretical prediction for the $W$-boson mass underscores the importance of conducting high-precision studies on the $W$ boson, which is one of the predominant objectives of proposed future $e^+e^-$ colliders. We investigate in detail the production of $W$-boson pairs at $e^+e^-$ colliders, and compute the next-to-next-to-leading order mixed QCD-EW corrections to both the integrated cross section and various kinematic distributions. By employing the method of differential equations, we analytically calculate the two-loop master integrals for the mixed QCD-EW virtual corrections to $e^+e^- \rightarrow W^+W^-$. Utilizing the Magnus transformation, we derive a set of canonical master integrals for each integral family. This canonical basis satisfies a system of differential equations in which the dependence on the dimensional regulator is linearly factorized from the kinematics. We then express all these canonical master integrals as Taylor series in $ε$ up to $ε^4$, with coefficients articulated in terms of Goncharov polylogarithms up to weight four. Upon applying our analytic expressions of these master integrals to the phenomenological analysis of $W$-pair production, we observe that the $\mathcal{O}(αα_s)$ corrections are significantly impactful in the $α(0)$ scheme, particularly in certain phase-space regions. However, these mixed QCD-EW corrections can be heavily suppressed by adopting the $G_μ$ scheme.

hep-ph↗

arXiv2024

AniGS: Animatable Gaussian Avatar from a Single Image with Inconsistent Gaussian Reconstruction

Generating animatable human avatars from a single image is essential for various digital human modeling applications. Existing 3D reconstruction methods often struggle to capture fine details in animatable models, while generative approaches for controllable animation, though avoiding explicit 3D modeling, suffer from viewpoint inconsistencies in extreme poses and computational inefficiencies. In this paper, we address these challenges by leveraging the power of generative models to produce detailed multi-view canonical pose images, which help resolve ambiguities in animatable human reconstruction. We then propose a robust method for 3D reconstruction of inconsistent images, enabling real-time rendering during inference. Specifically, we adapt a transformer-based video generation model to generate multi-view canonical pose images and normal maps, pretraining on a large-scale video dataset to improve generalization. To handle view inconsistencies, we recast the reconstruction problem as a 4D task and introduce an efficient 3D modeling approach using 4D Gaussian Splatting. Experiments demonstrate that our method achieves photorealistic, real-time animation of 3D human avatars from in-the-wild images, showcasing its effectiveness and generalization capability.

cs.CV↗

arXiv2024

A Compact Hybrid Battery Thermal Management System for Enhanced Cooling

Hybrid battery thermal management systems (HBTMS) combining active liquid cooling and passive phase change materials (PCM) cooling have shown a potential for the thermal management of lithium-ion batteries. However, the fill volume of coolant and PCM in hybrid cooling systems is limited by the size and weight of the HBTMS at high charge/discharge rates. These limitations result in reduced convective heat transfer from the coolant during discharge. The liquefaction rate of PCM is accelerated and the passive cooling effect is reduced. In this paper, we propose a compact hybrid cooling system with multi-inlet U-shaped microchannels for which the gap between channels is embedded by PCM/aluminum foam for compactness. Nanofluid cooling (NC) technology with better thermal conductivity is used. A pulsed flow function is further developed for enhanced cooling (EC) with reduced power consumption. An experimentally validated thermal-fluid dynamics model is developed to optimize operating conditions including coolant type, cooling direction, channel height, inlet flow rate, and cooling scheme. The results show that the hybrid cooling solution of NC+PCM+EC adopted by HBTMS further reduces the maximum temperature of the Li-ion battery by 3.44°C under a discharge rate of 1C at room temperature of 25°C with only a 5% increase in power consumption, compared to the conventional liquid cooling method for electric vehicles (EV). The average number of battery charges has increased by about 6 to 15 percent. The results of this study can help improve the range as well as driving safety of new energy EV.

eess.SY↗

arXiv2024

HHAvatar: Gaussian Head Avatar with Dynamic Hairs

Creating high-fidelity 3D head avatars has always been a research hotspot, but it remains a great challenge under lightweight sparse view setups. In this paper, we propose HHAvatar represented by controllable 3D Gaussians for high-fidelity head avatar with dynamic hair modeling. We first use 3D Gaussians to represent the appearance of the head, and then jointly optimize neutral 3D Gaussians and a fully learned MLP-based deformation field to capture complex expressions. The two parts benefit each other, thereby our method can model fine-grained dynamic details while ensuring expression accuracy. Furthermore, we devise a well-designed geometry-guided initialization strategy based on implicit SDF and Deep Marching Tetrahedra for the stability and convergence of the training procedure. To address the problem of dynamic hair modeling, we introduce a hybrid head model into our avatar representation based Gaussian Head Avatar and a training method that considers timing information and an occlusion perception module to model the non-rigid motion of hair. Experiments show that our approach outperforms other state-of-the-art sparse-view methods, achieving ultra high-fidelity rendering quality at 2K resolution even under exaggerated expressions and driving hairs reasonably with the motion of the head

cs.CV↗

arXiv2024

Two-loop planar master integrals for NNLO QCD corrections to W-pair production in quark-antiquark annihilation

The planar two-loop scalar Feynman integrals contributing to the massive NNLO QCD corrections for $W$-boson pair production via quark-antiquark annihilation can be classified into three family branches, each of which is reduced to a distinct set of master integrals (MIs), totaling $27$, $45$ and $15$, respectively. These MIs are analytically calculated using the method of differential equations, with solutions expanded as Taylor series in the dimensional regulator $ε$. For the first two family branches, the differential systems can be successfully transformed into canonical form by adopting appropriate bases of MIs. This enables the MIs of these family branches to be expressed either as Goncharov polylogarithms (GPLs) or as one-fold integrals over GPLs, up to $\mathcal{O}(ε^4)$. In contrast, the differential system for the third family branch can only be cast into a form linear in $ε$ due to the presence of elliptic integrals. The solution to this linear-form differential system is expressed in an iterated form owing to the strictly lower-triangular structure of the coefficient matrices at $ε= 0$. Our analytic expressions for these MIs are verified with high accuracy against the numerical results from the \texttt{AMFlow} package.

hep-ph↗

arXiv2024

Data-Efficient Generation for Dataset Distillation

While deep learning techniques have proven successful in image-related tasks, the exponentially increased data storage and computation costs become a significant challenge. Dataset distillation addresses these challenges by synthesizing only a few images for each class that encapsulate all essential information. Most current methods focus on matching. The problems lie in the synthetic images not being human-readable and the dataset performance being insufficient for downstream learning tasks. Moreover, the distillation time can quickly get out of bounds when the number of synthetic images per class increases even slightly. To address this, we train a class conditional latent diffusion model capable of generating realistic synthetic images with labels. The sampling time can be reduced to several tens of images per seconds. We demonstrate that models can be effectively trained using only a small set of synthetic images and evaluated on a large real test set. Our approach achieved rank $1$ in The First Dataset Distillation Challenge at ECCV 2024 on the CIFAR100 and TinyImageNet datasets.

cs.CV↗

arXiv2024

All Robots in One: A New Standard and Unified Dataset for Versatile, General-Purpose Embodied Agents

Embodied AI is transforming how AI systems interact with the physical world, yet existing datasets are inadequate for developing versatile, general-purpose agents. These limitations include a lack of standardized formats, insufficient data diversity, and inadequate data volume. To address these issues, we introduce ARIO (All Robots In One), a new data standard that enhances existing datasets by offering a unified data format, comprehensive sensory modalities, and a combination of real-world and simulated data. ARIO aims to improve the training of embodied AI agents, increasing their robustness and adaptability across various tasks and environments. Building upon the proposed new standard, we present a large-scale unified ARIO dataset, comprising approximately 3 million episodes collected from 258 series and 321,064 tasks. The ARIO standard and dataset represent a significant step towards bridging the gaps of existing data resources. By providing a cohesive framework for data collection and representation, ARIO paves the way for the development of more powerful and versatile embodied AI agents, capable of navigating and interacting with the physical world in increasingly complex and diverse ways. The project is available on https://imaei.github.io/project_pages/ario/

cs.RO↗

arXiv2024

Next-to-next-to-leading order $\text{QCD} \otimes \text{EW}$ corrections to $Z$-boson pair production at electron-positron colliders

We present a comprehensive analytic calculation of the next-to-next-to-leading order $\text{QCD} \otimes \text{EW}$ corrections to $Z$-boson pair production at electron-positron colliders. The two-loop master integrals essential to this calculation are evaluated using the differential equation method. In this work, we detail the formulation and solution of the canonical differential equations for the two-loop three-point master integrals with two on-shell $Z$-boson external legs and a massive internal quark in the loops. These canonical master integrals are systematically expanded as Taylor series in the dimensional regulator, $ε= (4-d)/2$, up to the order of $ε^4$, with coefficients expressed in terms of Goncharov polylogarithms up to weight four. Upon applying our analytic expressions of these master integrals to the phenomenological analysis of $Z$-pair production, we observe that the $\mathcal{O}(αα_s)$ corrections manifest at a level of approximately one percent compared to the leading-order predictions, underscoring their significance for comparisons with future high-precision experimental data.

hep-ph↗

arXiv2024

Orca: Ocean Significant Wave Height Estimation with Spatio-temporally Aware Large Language Models

Significant wave height (SWH) is a vital metric in marine science, and accurate SWH estimation is crucial for various applications, e.g., marine energy development, fishery, early warning systems for potential risks, etc. Traditional SWH estimation methods that are based on numerical models and physical theories are hindered by computational inefficiencies. Recently, machine learning has emerged as an appealing alternative to improve accuracy and reduce computational time. However, due to limited observational technology and high costs, the scarcity of real-world data restricts the potential of machine learning models. To overcome these limitations, we propose an ocean SWH estimation framework, namely Orca. Specifically, Orca enhances the limited spatio-temporal reasoning abilities of classic LLMs with a novel spatiotemporal aware encoding module. By segmenting the limited buoy observational data temporally, encoding the buoys' locations spatially, and designing prompt templates, Orca capitalizes on the robust generalization ability of LLMs to estimate significant wave height effectively with limited data. Experimental results on the Gulf of Mexico demonstrate that Orca achieves state-of-the-art performance in SWH estimation.

cs.LG↗

arXiv2024

Dreamer: Dual-RIS-aided Imager in Complementary Modes

Reconfigurable intelligent surfaces (RISs) have emerged as a promising auxiliary technology for radio frequency imaging. However, existing works face challenges of faint and intricate back-scattered waves and the restricted field-of-view (FoV), both resulting from complex target structures and a limited number of antennas. The synergistic benefits of multi-RIS-aided imaging hold promise for addressing these challenges. Here, we propose a dual-RIS-aided imaging system, Dreamer, which operates collaboratively in complementary modes (reflection-mode and transmission-mode). Dreamer significantly expands the FoV and enhances perception by deploying dual-RIS across various spatial and measurement patterns. Specifically, we perform a fine-grained analysis of how radio-frequency (RF) signals encode scene information in the scattered object modeling. Based on this modeling, we design illumination strategies to balance spatial resolution and observation scale, and implement a prototype system in a typical indoor environment. Moreover, we design a novel artificial neural network with a CNN-external-attention mechanism to translate RF signals into high-resolution images of human contours. Our approach achieves an impressive SSIM score exceeding 0.83, validating its effectiveness in broadening perception modes and enhancing imaging capabilities. The code to reproduce our results is available at https://github.com/fuhaiwang/Dreamer.

eess.SP↗

arXiv2024

MeshAvatar: Learning High-quality Triangular Human Avatars from Multi-view Videos

We present a novel pipeline for learning high-quality triangular human avatars from multi-view videos. Recent methods for avatar learning are typically based on neural radiance fields (NeRF), which is not compatible with traditional graphics pipeline and poses great challenges for operations like editing or synthesizing under different environments. To overcome these limitations, our method represents the avatar with an explicit triangular mesh extracted from an implicit SDF field, complemented by an implicit material field conditioned on given poses. Leveraging this triangular avatar representation, we incorporate physics-based rendering to accurately decompose geometry and texture. To enhance both the geometric and appearance details, we further employ a 2D UNet as the network backbone and introduce pseudo normal ground-truth as additional supervision. Experiments show that our method can learn triangular avatars with high-quality geometry reconstruction and plausible material decomposition, inherently supporting editing, manipulation or relighting operations.

cs.CV↗

arXiv2024

Image Distillation for Safe Data Sharing in Histopathology

Histopathology can help clinicians make accurate diagnoses, determine disease prognosis, and plan appropriate treatment strategies. As deep learning techniques prove successful in the medical domain, the primary challenges become limited data availability and concerns about data sharing and privacy. Federated learning has addressed this challenge by training models locally and updating parameters on a server. However, issues, such as domain shift and bias, persist and impact overall performance. Dataset distillation presents an alternative approach to overcoming these challenges. It involves creating a small synthetic dataset that encapsulates essential information, which can be shared without constraints. At present, this paradigm is not practicable as current distillation approaches only generate non human readable representations and exhibit insufficient performance for downstream learning tasks. We train a latent diffusion model and construct a new distilled synthetic dataset with a small number of human readable synthetic images. Selection of maximally informative synthetic images is done via graph community analysis of the representation space. We compare downstream classification models trained on our synthetic distillation data to models trained on real data and reach performances suitable for practical application.

cs.CV↗

arXiv2024

Deep learning quantum Monte Carlo for solids

Deep learning has deeply changed the paradigms of many research fields. At the heart of chemical and physical sciences is the accurate ab initio calculation of many-body wavefunction, which has become one of the most notable examples to demonstrate the power of deep learning in science. In particular, the introduction of deep learning into quantum Monte Carlo (QMC) has significantly advanced the frontier of ab initio calculation, offering a universal tool to solve the electronic structure of materials and molecules. Deep learning QMC architectures were initial designed and tested on small molecules, focusing on comparisons with other state-of-the-art ab initio methods. Methodological developments, including extensions to real solids and periodic models, have been rapidly progressing and reported applications are fast expanding. This review covers the theoretical foundation of deep learning QMC for solids, the neural network wavefunction ansatz, and various of other methodological developments. Applications on computing energy, electron density, electric polarization, force and stress of real solids are also reviewed. The methods have also been extended to other periodic systems and finite temperature calculations. The review highlights the potentials and existing challenges of deep learning QMC in materials chemistry and condensed matter physics.

physics.chem-ph↗

arXiv2024

PM-VIS+: High-Performance Video Instance Segmentation without Video Annotation

Video instance segmentation requires detecting, segmenting, and tracking objects in videos, typically relying on costly video annotations. This paper introduces a method that eliminates video annotations by utilizing image datasets. The PM-VIS algorithm is adapted to handle both bounding box and instance-level pixel annotations dynamically. We introduce ImageNet-bbox to supplement missing categories in video datasets and propose the PM-VIS+ algorithm to adjust supervision based on annotation types. To enhance accuracy, we use pseudo masks and semi-supervised optimization techniques on unannotated video data. This method achieves high video instance segmentation performance without manual video annotations, offering a cost-effective solution and new perspectives for video instance segmentation applications. The code will be available in https://github.com/ldknight/PM-VIS-plus

cs.CV↗

arXiv2024

Are AI-Generated Text Detectors Robust to Adversarial Perturbations?

The widespread use of large language models (LLMs) has sparked concerns about the potential misuse of AI-generated text, as these models can produce content that closely resembles human-generated text. Current detectors for AI-generated text (AIGT) lack robustness against adversarial perturbations, with even minor changes in characters or words causing a reversal in distinguishing between human-created and AI-generated text. This paper investigates the robustness of existing AIGT detection methods and introduces a novel detector, the Siamese Calibrated Reconstruction Network (SCRN). The SCRN employs a reconstruction network to add and remove noise from text, extracting a semantic representation that is robust to local perturbations. We also propose a siamese calibration technique to train the model to make equally confidence predictions under different noise, which improves the model's robustness against adversarial perturbations. Experiments on four publicly available datasets show that the SCRN outperforms all baseline methods, achieving 6.5\%-18.25\% absolute accuracy improvement over the best baseline method under adversarial attacks. Moreover, it exhibits superior generalizability in cross-domain, cross-genre, and mixed-source scenarios. The code is available at \url{https://github.com/CarlanLark/Robust-AIGC-Detector}.

cs.CL↗

arXiv2024

Multimechanism quantum anomalous Hall and Chern number tunable states in germanene (silicene, stanene)/$M$Bi$_2$Te$_4$ heterostructures

By constructing germanene (silicene, stanene)/$M$Bi$_2$Te$_4$ ($M$ = 3d-transition elements) heterostructures, we discovered and designed multimechanism quantum-anomalous-Hall (QAH) systems, including $Γ$-based QAH, $K$-$K'$-connected QAH, and valley-polarized $K$- or $K'$-based QAH states via first-principle computations. The unique systems possess a global gap and tunable Chern number. The coexisting conventional $Γ$-based QAH state of $M$Bi$_2$Te$_4$ and valley-polarized $K$($K'$)-based QAH state of germanene (silicene, stanene), with opposite chirality, can interact with each other. Adjusting magnetic configurations of $M$Bi$_2$Te$_4$-layers not only switch on (off) the QAH conductance, but also modulate Chern numbers exactly. For example, the germanene/bilayer-NiBi$_2$Te$_4$ possesses the Chern number $C = +1$ in ferromagnetic couplings and $C = +2$ in antiferromagnetic couplings. The novel multimechanism QAH insulators, which are achievable in experiments, provide a new approach to spintronics and valleytronics based on topological states of matter.

cond-mat.mtrl-sci↗