arXiv subjects

Xiaoye Liang

Publications and source records attributed to Xiaoye Liang.

6 recordsLinked to original sources

arXiv2026

FDM-MFVT: Few-step Sampling Diffusion Model for Mask-Free Virtual Try-On

Image-based Virtual Try-On (IVTON) has greatly advanced through diffusion models, yet existing methods require many sampling steps and depend on masks with costly auxiliary networks. In addition, the absence of large-scale mask-free paired datasets further limits the development of mask-free IVTON. We propose FDM-MFVT, a few-step diffusion model for mask-free IVTON, integrating an Outfit-aware Noise Optimization Module (OANO) and an Instruction-driven Try-on Module (IDT) to enhance efficiency and flexibility.The OANO module initializes the alignment space with noise using the input image and only needs 6 steps to generate a higher-fidelity try-on image compared to 30 steps.The IDT module uses virtual try-on prompts and efficient adaptation to generate high-quality results from garment and person images alone. We further introduce MFVT, a 30,000-pair mask-free IVTON dataset. Experiments show that FDM-MFVT achieves superior quantitative and qualitative results with fewer inference steps than mask-based and mask-free baseline methods.

cs.CV↗

arXiv2026

VTEdit-Bench: A Comprehensive Benchmark for Multi-Reference Image Editing Models in Virtual Try-On

As virtual try-on (VTON) continues to advance, a growing number of real-world scenarios have emerged, pushing beyond the ability of the existing specialized VTON models. Meanwhile, universal multi-reference image editing models have progressed rapidly and exhibit strong generalization in visual editing, suggesting a promising route toward more flexible VTON systems. However, despite their strong capabilities, the strengths and limitations of universal editors for VTON remain insufficiently explored due to the lack of systematic evaluation benchmarks. To address this gap, we introduce VTEdit-Bench, a comprehensive benchmark designed to evaluate universal multi-reference image editing models across various realistic VTON scenarios. VTEdit-Bench contains 24,220 test image pairs spanning five representative VTON tasks with progressively increasing complexity, enabling systematic analysis of robustness and generalization. We further propose VTEdit-QA, a reference-aware VLM-based evaluator that assesses VTON performance from three key aspects: model consistency, cloth consistency, and overall image quality. Through this framework, we systematically evaluate eight universal editing models and compare them with seven specialized VTON models. Results show that top universal editors are competitive on conventional tasks and generalize more stably to harder scenarios, but remain challenged by complex reference configurations, particularly multi-cloth conditioning.

cs.CV↗

arXiv2025

Burst Image Quality Assessment: A New Benchmark and Unified Framework for Multiple Downstream Tasks

In recent years, the development of burst imaging technology has improved the capture and processing capabilities of visual data, enabling a wide range of applications. However, the redundancy in burst images leads to the increased storage and transmission demands, as well as reduced efficiency of downstream tasks. To address this, we propose a new task of Burst Image Quality Assessment (BuIQA), to evaluate the task-driven quality of each frame within a burst sequence, providing reasonable cues for burst image selection. Specifically, we establish the first benchmark dataset for BuIQA, consisting of $7,346$ burst sequences with $45,827$ images and $191,572$ annotated quality scores for multiple downstream scenarios. Inspired by the data analysis, a unified BuIQA framework is proposed to achieve an efficient adaption for BuIQA under diverse downstream scenarios. Specifically, a task-driven prompt generation network is developed with heterogeneous knowledge distillation, to learn the priors of the downstream task. Then, the task-aware quality assessment network is introduced to assess the burst image quality based on the task prompt. Extensive experiments across 10 downstream scenarios demonstrate the impressive BuIQA performance of the proposed approach, outperforming the state-of-the-art. Furthermore, it can achieve $0.33$ dB PSNR improvement in the downstream tasks of denoising and super-resolution, by applying our approach to select the high-quality burst frames.

cs.CV↗

arXiv2023

On the (non-)existence of tight distance-regular graphs: a local approach

Let $\Gamma$ denote a distance-regular graph with diameter $D\geq 3$. Juri\v{s}i\'c and Vidali conjectured that if $\Gamma$ is tight with classical parameters $(D,b,\alpha,\beta)$, $b\geq 2$, then $\Gamma$ is not locally the block graph of an orthogonal array nor the block graph of a Steiner system. In the present paper, we prove this conjecture and, furthermore, extend it from the following aspect. Assume that for every triple of vertices $x, y, z$ of $\Gamma$, where $x$ and $y$ are adjacent, and $z$ is at distance $2$ from both $x$ and $y$, the number of common neighbors of $x$, $y$, $z$ is constant. We then show that if $\Gamma$ is locally the block graph of an orthogonal array (resp. a Steiner system) with smallest eigenvalue $-m$, $m\geq 3$, then the intersection number $c_2$ is not equal to $m^2$ (resp. $m(m+1)$). Using this result, we prove that if a tight distance-regular graph $\Gamma$ is not locally the block graph of an orthogonal array or a Steiner system, then the valency (and hence diameter) of $\Gamma$ is bounded by a function in the parameter $b=b_1/(1+\theta_1)$, where $b_1$ is the intersection number of $\Gamma$ and $\theta_1$ is the second largest eigenvalue of $\Gamma$.

math.CO↗

arXiv2023

Non-geometric distance-regular graphs of diameter at least $3$ with smallest eigenvalue at least $-3$

In this paper, we classify non-geometric distance-regular graphs of diameter at least $3$ with smallest eigenvalue at least $-3$. This is progress towards what is hoped to be an eventual complete classification of distance-regular graphs with smallest eigenvalue at least $-3$, analogous to existing classification results available in the case that the smallest eigenvalue is at least $-2$.

math.CO↗

arXiv2021

A duality of scaffolds for translation association schemes

Scaffolds are certain tensors arising in the study of association schemes, and have been (implicitly) understood diagrammatically as digraphs with distinguished "root" nodes and with matrix edge weights, often taken from Bose-Mesner algebras. In this paper, we first present a slight modification of Martin's conjecture (2021) concerning a duality of scaffolds whose digraphs are embedded in a closed disk in the plane with root nodes all lying on the boundary circle, and then show that this modified conjecture holds true if we restrict ourselves to the class of translation association schemes, i.e., those association schemes that admit abelian regular automorphism groups.

math.CO↗