arXiv subjects

Yupeng Zhu

Publications and source records attributed to Yupeng Zhu.

8 recordsLinked to original sources

arXiv2026

World Narrative Model for Highly Controllable Video Generation: A Paradigm Shift from Pixel Sampling to Physical World Orchestration

The fundamental obstacle to industrial grade video generation is the lack of controllability: existing models treat video as a pixel distribution sampling problem, bypassing the explicit, instance level $4D$ $(3D + T)$ physical world. Consequently, content creators cannot specify geometry, motion, camera parameters, or lighting in a deterministic, quantitative way, leading to the infamous ''gacha'' loop that makes professional content creation prohibitively inefficient and expensive. To address this, we introduce the World Narrative Model (WNM), a paradigm that decouples what to render -- the structured physical narrative -- from how to render -- the pixel generation process. WNM replaces end-to-end black-box sampling with orchestrated $4D$ pre-visualization for media generation. Collaborative agents translate sparse multimodal inputs, including text, reference videos, and sketches, into a fully editable world representation with scene geometry, object layouts, character/animal skeleton motion, trajectories, camera motion, and lighting at quantitative, physically meaningful granularity. This representation acts as a deterministic structural blueprint that drives existing video foundation models, either frozen or lightly adapted, to render final footage, turning the base model into a faithful neural shader. Built on this engine, our human-AI platform supports automatic world generation and pre-visualization aligned with professional filmmaking pipelines, while director consoles enable seamless human refinement. Experiments show that WNM greatly reduces probabilistic ``gacha'' calls and produces videos whose layout, motion, and cinematography closely follow creator intent. The framework is open and modular, allowing each component, such as world representation, control agents, and adapters, to be independently improved. Project website: https://glassroom.sjtu.edu.cn/WNM/.

cs.CV↗

arXiv2026

ProxyImg: Towards Highly-Controllable Image Representation via Hierarchical Disentangled Proxy Embedding

Prevailing image representation methods, including explicit representations such as raster images and Gaussian primitives, as well as implicit representations such as latent images, either suffer from representation redundancy that leads to heavy manual editing effort, or lack a direct mapping from latent variables to semantic instances or parts, making fine-grained manipulation difficult. These limitations hinder efficient and controllable image and video editing. To address these issues, we propose a hierarchical proxy-based parametric image representation that disentangles semantic, geometric, and textural attributes into independent and manipulable parameter spaces. Based on a semantic-aware decomposition of the input image, our representation constructs hierarchical proxy geometries through adaptive Bezier fitting and iterative internal region subdivision and meshing. Multi-scale implicit texture parameters are embedded into the resulting geometry-aware distributed proxy nodes, enabling continuous high-fidelity reconstruction in the pixel domain and instance- or part-independent semantic editing. In addition, we introduce a locality-adaptive feature indexing mechanism to ensure spatial texture coherence, which further supports high-quality background completion without relying on generative models. Extensive experiments on image reconstruction and editing benchmarks, including ImageNet, OIR-Bench, and HumanEdit, demonstrate that our method achieves state-of-the-art rendering fidelity with significantly fewer parameters, while enabling intuitive, interactive, and physically plausible manipulation. Moreover, by integrating proxy nodes with Position-Based Dynamics, our framework supports real-time physics-driven animation using lightweight implicit rendering, achieving superior temporal consistency and visual realism compared with generative approaches.

cs.CV↗

arXiv2025

3DProxyImg: Controllable 3D-Aware Animation Synthesis from Single Image via 2D-3D Aligned Proxy Embedding

3D animation is central to modern visual media, yet traditional production pipelines remain labor-intensive, expertise-demanding, and computationally expensive. Recent AIGC-based approaches partially automate asset creation and rigging, but they either inherit the heavy costs of full 3D pipelines or rely on video-synthesis paradigms that sacrifice 3D controllability and interactivity. We focus on single-image 3D animation generation and argue that progress is fundamentally constrained by a trade-off between rendering quality and 3D control. To address this limitation, we propose a lightweight 3D animation framework that decouples geometric control from appearance synthesis. The core idea is a 2D-3D aligned proxy representation that uses a coarse 3D estimate as a structural carrier, while delegating high-fidelity appearance and view synthesis to learned image-space generative priors. This proxy formulation enables 3D-aware motion control and interaction comparable to classical pipelines, without requiring accurate geometry or expensive optimization, and naturally extends to coherent background animation. Extensive experiments demonstrate that our method achieves efficient animation generation on low-power platforms and outperforms video-based 3D animation generation in identity preservation, geometric and textural consistency, and the level of precise, interactive control it offers to users.

cs.CV↗

arXiv2025

Vectorized Video Representation with Easy Editing via Hierarchical Spatio-Temporally Consistent Proxy Embedding

Current video representations heavily rely on unstable and over-grained priors for motion and appearance modelling, \emph{i.e.}, pixel-level matching and tracking. A tracking error of just a few pixels would lead to the collapse of the visual object representation, not to mention occlusions and large motion frequently occurring in videos. To overcome the above mentioned vulnerability, this work proposes spatio-temporally consistent proxy nodes to represent dynamically changing objects/scenes in the video. On the one hand, the hierarchical proxy nodes have the ability to stably express the multi-scale structure of visual objects, so they are not affected by accumulated tracking error, long-term motion, occlusion, and viewpoint variation. On the other hand, the dynamic representation update mechanism of the proxy nodes adequately leverages spatio-temporal priors of the video to mitigate the impact of inaccurate trackers, thereby effectively handling drastic changes in scenes and objects. Additionally, the decoupled encoding manner of the shape and texture representations across different visual objects in the video facilitates controllable and fine-grained appearance editing capability. Extensive experiments demonstrate that the proposed representation achieves high video reconstruction accuracy with fewer parameters and supports complex video processing tasks, including video in-painting and keyframe-based temporally consistent video editing.

cs.CV↗

arXiv2025

QoE-oriented Communication Service Provision for Annotation Rendering in Mobile Augmented Reality

As mobile augmented reality (MAR) continues to evolve, future 6G networks will play a pivotal role in supporting immersive and personalized user experiences. In this paper, we address the communication service provision problem for annotation rendering in edge-assisted MAR, with the objective of optimizing spectrum resource utilization while ensuring the required quality of experience (QoE) for MAR users. To overcome the challenges of user-specific uplink data traffic patterns and the complex operational mechanisms of annotation rendering, we propose a digital twin (DT)-based approach. We first design a DT specifically tailored for MAR applications to learn key annotation rendering mechanisms, enabling the network controller to access MAR application-specific information. Then, we develop a DT based QoE modeling approach to capture the unique relationship between individual user QoE and spectrum resource demands. Finally, we propose a QoE-oriented resource allocation algorithm that decreases resource utilization compared to conventional net work slicing-based approaches. Simulation results demonstrate that our DT-based approach outperforms benchmark approaches in the accuracy and granularity of QoE modeling.

eess.IV↗

arXiv2024

Toward Tiny and High-quality Facial Makeup with Data Amplify Learning

Contemporary makeup approaches primarily hinge on unpaired learning paradigms, yet they grapple with the challenges of inaccurate supervision (e.g., face misalignment) and sophisticated facial prompts (including face parsing, and landmark detection). These challenges prohibit low-cost deployment of facial makeup models, especially on mobile devices. To solve above problems, we propose a brand-new learning paradigm, termed "Data Amplify Learning (DAL)," alongside a compact makeup model named "TinyBeauty." The core idea of DAL lies in employing a Diffusion-based Data Amplifier (DDA) to "amplify" limited images for the model training, thereby enabling accurate pixel-to-pixel supervision with merely a handful of annotations. Two pivotal innovations in DDA facilitate the above training approach: (1) A Residual Diffusion Model (RDM) is designed to generate high-fidelity detail and circumvent the detail vanishing problem in the vanilla diffusion models; (2) A Fine-Grained Makeup Module (FGMM) is proposed to achieve precise makeup control and combination while retaining face identity. Coupled with DAL, TinyBeauty necessitates merely 80K parameters to achieve a state-of-the-art performance without intricate face prompts. Meanwhile, TinyBeauty achieves a remarkable inference speed of up to 460 fps on the iPhone 13. Extensive experiments show that DAL can produce highly competitive makeup models using only 5 image pairs.

cs.CV↗

arXiv2023

Single crystal synthesis and low-lying electronic structure of V$_3$S$_4$

We report successful growth of millimeter-sized high quality single crystals of V$_3$S$_4$, a candidate topological semimetal belonging to a low-symmetry space group and consisting of only low atomic number elements. Using density functional theory calculations and angle-resolved photoemission spectroscopy, we show that the nonmagnetic phase of monoclinic V$_3$S$_4$ hosts type-II Dirac-like quasiparticles which opens a sizable gap due to spin orbit coupling, as well as theoretical multiple nodal lines that are eliminated also by spin orbit coupling. These results suggest that relativistic effects give rise to profound modifications of the topological properties even in compounds with low-weight elements.

cond-mat.mtrl-sci↗

arXiv2020

Ultra-sensitive nanometric flat pigment for binocular stereoscopic image

Two-dimensional (2D) transition metal dichalcogenides (TMDs) with tantalizing layer-dependent electronic and optical properties have emerged as a new paradigm for integrated flat opto-electronic devices. However, daunting challenges remain in deterministic fabrication of TMD layers with demanded shapes and thicknesses as well as light field manipulation in such atomic-thick layers with vanishingly small thicknesses compared to the wavelength. Here, we demonstrate ultra-sensitive light field manipulation in full visible ranges based on laser exfoliating MoS2 layers with nanometric precisions. The nontrivial interfacial phase shifts stemming from the unique dispersion of MoS2 layers integrated on the metallic substrate empower an ultra-sensitive resonance manipulation up to 12.8 nm per MoS2 layer across the entire visible bands, which is more than five times larger than their counterparts. The interlayer van der Waals interactions endow a laser exfoliation method for on-demand patterning MoS2 with atomic thickness precisions and subwavelength feature sizes in a facile and lithography-free fashion. With this, nanometric flat color prints and further binocular stereoscopic views by multi-perspective diffractive images can be realized. Our results with demonstrated practicality unlock full potentials and pave the way for widespread applications of emerging 2D flat optics.

physics.optics↗