arXiv subjects

Ziyu Zhang

Publications and source records attributed to Ziyu Zhang.

At least 55 records · Page 3Linked to original sources

arXiv2025

Deformations of differential equations

We study perturbations of linear differential equations, deriving explicit series solutions, using Dyson-type expansions. We analyze the monodromy of deformed solutions in a number of examples, and relate this to cocycles in a cohomological framework. We also analyze the spectral properties of the hypergeometric equation under infinitesimal deformations, derive the first-order eigenvalue correction via orthogonality, and establish natural bounds for the perturbed problem.

math.CA↗

arXiv2025

Quadratic Gaussian Splatting: High Quality Surface Reconstruction with Second-order Geometric Primitives

We propose Quadratic Gaussian Splatting (QGS), a novel representation that replaces static primitives with deformable quadric surfaces (e.g., ellipse, paraboloids) to capture intricate geometry. Unlike prior works that rely on Euclidean distance for primitive density modeling--a metric misaligned with surface geometry under deformation--QGS introduces geodesic distance-based density distributions. This innovation ensures that density weights adapt intrinsically to the primitive curvature, preserving consistency during shape changes (e.g., from planar disks to curved paraboloids). By solving geodesic distances in closed form on quadric surfaces, QGS enables surface-aware splatting, where a single primitive can represent complex curvature that previously required dozens of planar surfels, potentially reducing memory usage while maintaining efficient rendering via fast ray-quadric intersection. Experiments on DTU, Tanks and Temples, and MipNeRF360 datasets demonstrate state-of-the-art surface reconstruction, with QGS reducing geometric error (chamfer distance) by 33% over 2DGS and 27% over GOF on the DTU dataset. Crucially, QGS retains competitive appearance quality, bridging the gap between geometric precision and visual fidelity for applications like robotics and immersive reality.

cs.CV↗

arXiv2025

Narrative-Driven Travel Planning: Geoculturally-Grounded Script Generation with Evolutionary Itinerary Optimization

To enhance tourists' experiences and immersion, this paper proposes a narrative-driven travel planning framework called NarrativeGuide, which generates a geoculturally-grounded narrative script for travelers, offering a novel, role-playing experience for their journey. In the initial stage, NarrativeGuide constructs a knowledge graph for attractions within a city, then configures the worldview, character setting, and exposition based on the knowledge graph. Using this foundation, the knowledge graph is combined to generate an independent scene unit for each attraction. During the itinerary planning stage, NarrativeGuide models narrative-driven travel planning as an optimization problem, utilizing a genetic algorithm (GA) to refine the itinerary. Before evaluating the candidate itinerary, transition scripts are generated for each pair of adjacent attractions, which, along with the scene units, form a complete script. The weighted sum of script coherence, travel time, and attraction scores is then used as the fitness value to update the candidate solution set. In our experiments, we incorporated the TravelPlanner benchmark to systematically evaluate the planning capability of NarrativeGuide under complex constraints. In addition, we assessed its performance in terms of narrative coherence and cultural fit. The results show that NarrativeGuide demonstrates strong capabilities in both itinerary planning and script generation.

cs.AI↗

arXiv2025

Pointing the Way: Refining Radar-Lidar Localization Using Learned ICP Weights

This paper presents a novel deep-learning-based approach to improve localizing radar measurements against lidar maps. This radar-lidar localization leverages the benefits of both sensors; radar is resilient against adverse weather, while lidar produces high-quality maps in clear conditions. However, owing in part to the unique artefacts present in radar measurements, radar-lidar localization has struggled to achieve comparable performance to lidar-lidar systems, preventing it from being viable for autonomous driving. This work builds on ICP-based radar-lidar localization by including a learned preprocessing step that weights radar points based on high-level scan information. To train the weight-generating network, we present a novel, stand-alone, open-source differentiable ICP library. The learned weights facilitate ICP by filtering out harmful radar points related to artefacts, noise, and even vehicles on the road. Combining an analytical approach with a learned weight reduces overall localization errors and improves convergence in radar-lidar ICP results run on real-world autonomous driving data. Our code base is publicly available to facilitate reproducibility and extensions.

cs.RO↗

arXiv2025

Decoding Speaker-Normalized Pitch from EEG for Mandarin Perception

The same speech content produced by different speakers exhibits significant differences in pitch contour, yet listeners' semantic perception remains unaffected. This phenomenon may stem from the brain's perception of pitch contours being independent of individual speakers' pitch ranges. In this work, we recorded electroencephalogram (EEG) while participants listened to Mandarin monosyllables with varying tones, phonemes, and speakers. The CE-ViViT model is proposed to decode raw or speaker-normalized pitch contours directly from EEG. Experimental results demonstrate that the proposed model can decode pitch contours with modest errors, achieving performance comparable to state-of-the-art EEG regression methods. Moreover, speaker-normalized pitch contours were decoded more accurately, supporting the neural encoding of relative pitch.

cs.SD↗

arXiv2025

Spark-TTS: An Efficient LLM-Based Text-to-Speech Model with Single-Stream Decoupled Speech Tokens

Recent advancements in large language models (LLMs) have driven significant progress in zero-shot text-to-speech (TTS) synthesis. However, existing foundation models rely on multi-stage processing or complex architectures for predicting multiple codebooks, limiting efficiency and integration flexibility. To overcome these challenges, we introduce Spark-TTS, a novel system powered by BiCodec, a single-stream speech codec that decomposes speech into two complementary token types: low-bitrate semantic tokens for linguistic content and fixed-length global tokens for speaker attributes. This disentangled representation, combined with the Qwen2.5 LLM and a chain-of-thought (CoT) generation approach, enables both coarse-grained control (e.g., gender, speaking style) and fine-grained adjustments (e.g., precise pitch values, speaking rate). To facilitate research in controllable TTS, we introduce VoxBox, a meticulously curated 100,000-hour dataset with comprehensive attribute annotations. Extensive experiments demonstrate that Spark-TTS not only achieves state-of-the-art zero-shot voice cloning but also generates highly customizable voices that surpass the limitations of reference-based synthesis. Source code, pre-trained models, and audio samples are available at https://github.com/SparkAudio/Spark-TTS.

cs.SD↗

arXiv2024

The NPU-HWC System for the ISCSLP 2024 Inspirational and Convincing Audio Generation Challenge

This paper presents the NPU-HWC system submitted to the ISCSLP 2024 Inspirational and Convincing Audio Generation Challenge 2024 (ICAGC). Our system consists of two modules: a speech generator for Track 1 and a background audio generator for Track 2. In Track 1, we employ Single-Codec to tokenize the speech into discrete tokens and use a language-model-based approach to achieve zero-shot speaking style cloning. The Single-Codec effectively decouples timbre and speaking style at the token level, reducing the acoustic modeling burden on the autoregressive language model. Additionally, we use DSPGAN to upsample 16 kHz mel-spectrograms to high-fidelity 48 kHz waveforms. In Track 2, we propose a background audio generator based on large language models (LLMs). This system produces scene-appropriate accompaniment descriptions, synthesizes background audio with Tango 2, and integrates it with the speech generated by our Track 1 system. Our submission achieves the second place and the first place in Track 1 and Track 2 respectively.

cs.SD↗

arXiv2024

A smooth but non-symplectic moduli of sheaves on a hyperkähler variety

For an abelian surface $A$, we consider stable vector bundles on a generalized Kummer variety $K_n(A)$ with $n>1$. We prove that the connected component of the moduli space which contains the tautological bundles associated to line bundles of degree $0$ is isomorphic to the blowup of the dual abelian surface in one point. We believe that this is the first explicit example of a component which is smooth with a non-trivial canonical bundle.

math.AG↗

arXiv2023

Near-Linear Time and Fixed-Parameter Tractable Algorithms for Tensor Decompositions

We study low rank approximation of tensors, focusing on the tensor train and Tucker decompositions, as well as approximations with tree tensor networks and more general tensor networks. For tensor train decomposition, we give a bicriteria $(1 + \eps)$-approximation algorithm with a small bicriteria rank and $O(q \cdot \nnz(A))$ running time, up to lower order terms, which improves over the additive error algorithm of \cite{huber2017randomized}. We also show how to convert the algorithm of \cite{huber2017randomized} into a relative error algorithm, but their algorithm necessarily has a running time of $O(qr^2 \cdot \nnz(A)) + n \cdot \poly(qk/\eps)$ when converted to a $(1 + \eps)$-approximation algorithm with bicriteria rank $r$. To the best of our knowledge, our work is the first to achieve polynomial time relative error approximation for tensor train decomposition. Our key technique is a method for obtaining subspace embeddings with a number of rows polynomial in $q$ for a matrix which is the flattening of a tensor train of $q$ tensors. We extend our algorithm to tree tensor networks. In addition, we extend our algorithm to tensor networks with arbitrary graphs (which we refer to as general tensor networks), by using a result of \cite{ms08_simulating_quantum_tensor_contraction} and showing that a general tensor network of rank $k$ can be contracted to a binary tree network of rank $k^{O(°(G)\tw(G))}$, allowing us to reduce to the case of tree tensor networks. Finally, we give new fixed-parameter tractable algorithms for the tensor train, Tucker, and CP decompositions, which are simpler than those of \cite{swz19_tensor_low_rank} since they do not make use of polynomial system solvers. Our technique of Gaussian subspace embeddings with exactly $k$ rows (and thus exponentially small success probability) may be of independent interest.

cs.DS↗

arXiv2023

SST-ReversibleNet: Reversible-prior-based Spectral-Spatial Transformer for Efficient Hyperspectral Image Reconstruction

Spectral image reconstruction is an important task in snapshot compressed imaging. This paper aims to propose a new end-to-end framework with iterative capabilities similar to a deep unfolding network to improve reconstruction accuracy, independent of optimization conditions, and to reduce the number of parameters. A novel framework called the reversible-prior-based method is proposed. Inspired by the reversibility of the optical path, the reversible-prior-based framework projects the reconstructions back into the measurement space, and then the residuals between the projected data and the real measurements are fed into the network for iteration. The reconstruction subnet in the network then learns the mapping of the residuals to the true values to improve reconstruction accuracy. In addition, a novel spectral-spatial transformer is proposed to account for the global correlation of spectral data in both spatial and spectral dimensions while balancing network depth and computational complexity, in response to the shortcomings of existing transformer-based denoising modules that ignore spatial texture features or learn local spatial features at the expense of global spatial features. Extensive experiments show that our SST-ReversibleNet significantly outperforms state-of-the-art methods on simulated and real HSI datasets, while requiring lower computational and storage costs. https://github.com/caizeyu1992/SST

eess.IV↗

arXiv2022

Efficient Single-Image Depth Estimation on Mobile Devices, Mobile AI & AIM 2022 Challenge: Report

Various depth estimation models are now widely used on many mobile and IoT devices for image segmentation, bokeh effect rendering, object tracking and many other mobile tasks. Thus, it is very crucial to have efficient and accurate depth estimation models that can run fast on low-power mobile chipsets. In this Mobile AI challenge, the target was to develop deep learning-based single image depth estimation solutions that can show a real-time performance on IoT platforms and smartphones. For this, the participants used a large-scale RGB-to-depth dataset that was collected with the ZED stereo camera capable to generated depth maps for objects located at up to 50 meters. The runtime of all models was evaluated on the Raspberry Pi 4 platform, where the developed solutions were able to generate VGA resolution depth maps at up to 27 FPS while achieving high fidelity results. All models developed in the challenge are also compatible with any Android or Linux-based mobile devices, their detailed description is provided in this paper.

cs.CV↗

arXiv2022

Stable vector bundles on generalized Kummer varieties

For an abelian surface $A$, we explicitly construct two new families of stable vector bundles on the generalized Kummer variety $K_n(A)$ for $n\geqslant 2$. The first is the family of tautological bundles associated to stable bundles on $A$, and the second is the family of the "wrong-way" fibers of a universal family of stable bundles on the dual abelian surface $\widehat{A}$ parametrized by $K_n(A)$. Each family exhibits a smooth connected component in the moduli space of stable bundles on $K_n(A)$, which is holomorphic symplectic but not simply connected, contrary to the case of K3 surfaces.

math.AG↗

arXiv2021

Relative VGIT and an application to degenerations of Hilbert schemes

We generalize the classical semi-continuity theorem for GIT (semi)stable loci under variations of linearizations to a relative situation of an equivariant projective morphism from X to an affine base S. As an application to moduli problems, we consider degenerations of Hilbert schemes, and give a conceptual interpretation of the (semi)stable loci of the degeneration families constructed by Gulbrandsen-Halle-Hulek.

math.AG↗

arXiv2021

Stability of some vector bundles on Hilbert schemes of points on K3 surfaces

Let $X$ be a projective K3 surfaces. In two examples where there exists a fine moduli space $M$ of stable vector bundles on $X$, isomorphic to a Hilbert scheme of points, we prove that the universal family $\mathcal{E}$ on $X\times M$ can be understood as a complete flat family of stable vector bundles on $M$ parametrized by $X$, which identifies $X$ with a smooth connected component of some moduli space of stable sheaves on $M$.

math.AG↗

arXiv2021

Using Deep Learning Sequence Models to Identify SARS-CoV-2 Divergence

SARS-CoV-2 is an upper respiratory system RNA virus that has caused over 3 million deaths and infecting over 150 million worldwide as of May 2021. With thousands of strains sequenced to date, SARS-CoV-2 mutations pose significant challenges to scientists on keeping pace with vaccine development and public health measures. Therefore, an efficient method of identifying the divergence of lab samples from patients would greatly aid the documentation of SARS-CoV-2 genomics. In this study, we propose a neural network model that leverages recurrent and convolutional units to directly take in amino acid sequences of spike proteins and classify corresponding clades. We also compared our model's performance with Bidirectional Encoder Representations from Transformers (BERT) pre-trained on protein database. Our approach has the potential of providing a more computationally efficient alternative to current homology based intra-species differentiation.

q-bio.QM↗

arXiv2021

Fast and Accurate Single-Image Depth Estimation on Mobile Devices, Mobile AI 2021 Challenge: Report

Depth estimation is an important computer vision problem with many practical applications to mobile devices. While many solutions have been proposed for this task, they are usually very computationally expensive and thus are not applicable for on-device inference. To address this problem, we introduce the first Mobile AI challenge, where the target is to develop an end-to-end deep learning-based depth estimation solutions that can demonstrate a nearly real-time performance on smartphones and IoT platforms. For this, the participants were provided with a new large-scale dataset containing RGB-depth image pairs obtained with a dedicated stereo ZED camera producing high-resolution depth maps for objects located at up to 50 meters. The runtime of all models was evaluated on the popular Raspberry Pi 4 platform with a mobile ARM-based Broadcom chipset. The proposed solutions can generate VGA resolution depth maps at up to 10 FPS on the Raspberry Pi 4 while achieving high fidelity results, and are compatible with any Android or Linux-based mobile devices. A detailed description of all models developed in the challenge is provided in this paper.

eess.IV↗

arXiv2021

Few-Data Guided Learning Upon End-to-End Point Cloud Network for 3D Face Recognition

3D face recognition has shown its potential in many application scenarios. Among numerous 3D face recognition methods, deep-learning-based methods have developed vigorously in recent years. In this paper, an end-to-end deep learning network entitled Sur3dNet-Face for point-cloud-based 3D face recognition is proposed. The network uses PointNet as the backbone, which is a successful point cloud classification solution but does not work properly in face recognition. Supplemented with modifications in network architecture and a few-data guided learning framework based on Gaussian process morphable model, the backbone is successfully modified for 3D face recognition. Different from existing methods training with a large amount of data in multiple datasets, our method uses Spring2003 subset of FRGC v2.0 for training which contains only 943 facial scans, and the network is well trained with the guidance of such a small amount of real data. Without fine-tuning on the test set, the Rank-1 Recognition Rate (RR1) is achieved as follows: 98.85% on FRGC v2.0 dataset and 99.33% on Bosphorus dataset, which proves the effectiveness and the potentiality of our method.

cs.CV↗

arXiv2020

On the motive of O'Grady's ten-dimensional hyper-Kähler varieties

We investigate how the motive of hyper-Kähler varieties is controlled by weight-2 (or surface-like) motives via tensor operations. In the first part, we study the Voevodsky motive of singular moduli spaces of semistable sheaves on K3 and abelian surfaces as well as the Chow motive of their crepant resolutions, when they exist. We show that these motives are in the tensor subcategory generated by the motive of the surface, provided that a crepant resolution exists. This extends a recent result of Bülles to the O'Grady-10 situation. In the non-commutative setting, similar results are proved for the Chow motive of moduli spaces of stable objects of the K3 category of a cubic fourfold. As a consequence, we provide abundant examples of hyper-Kähler varieties of O'Grady-10 deformation type satisfying the standard conjectures. In the second part, we study the André motive of projective hyper-Kähler varieties. We attach to any such variety its defect group, an algebraic group which acts on the cohomology and measures the difference between the full motive and its weight-2 part. When the second Betti number is not 3, we show that the defect group is a natural complement of the Mumford--Tate group inside the motivic Galois group, and that it is deformation invariant. We prove the triviality of this group for all known examples of projective hyper-Kähler varieties, so that in each case the full motive is controlled by its weight-2 part. As applications, we show that for any variety motivated by a product of known hyper-Kähler varieties, all Hodge and Tate classes are motivated, the motivated Mumford--Tate conjecture holds, and the André motive is abelian. This last point completes a recent work of Soldatenkov and provides a different proof for some of his results.

math.AG↗