arXiv subjects

Zhuoqian Yang

Publications and source records attributed to Zhuoqian Yang.

10 recordsLinked to original sources

arXiv2026

Weight Space Representation Learning via Neural Field Adaptation

We investigate the potential of weights to serve as effective representations, focusing on neural fields. Our key insight is that constraining the optimization space through a pre-trained base model and low-rank adaptation (LoRA) can induce structure in weight space. Across reconstruction, generation, and analysis tasks on 2D and 3D data, we find that multiplicative LoRA weights achieve high representation quality while exhibiting distinctiveness and semantic structure. When used with latent diffusion models, multiplicative LoRA weights enable higher-quality generation than existing weight-space methods.

cs.LG↗

arXiv2026

Subtractive Modulative Network with Learnable Periodic Activations

We propose the Subtractive Modulative Network (SMN), a novel, parameter-efficient Implicit Neural Representation (INR) architecture inspired by classical subtractive synthesis. The SMN is designed as a principled signal processing pipeline, featuring a learnable periodic activation layer (Oscillator) that generates a multi-frequency basis, and a series of modulative mask modules (Filters) that actively generate high-order harmonics. We provide both theoretical analysis and empirical validation for our design. Our SMN achieves a PSNR of $40+$ dB on two image datasets, comparing favorably against state-of-the-art methods in terms of both reconstruction accuracy and parameter efficiency. Furthermore, consistent advantage is observed on the challenging 3D NeRF novel view synthesis task. Supplementary materials are available at https://inrainbws.github.io/smn/.

cs.CV↗

arXiv2024

Integrating optimal ridesharing matching into multimodal traffic model: Implications for policy and sustainable transport system

Integrating ridesharing matching explicitly into multimodal traffic models is crucial for accurately assessing the impacts of multimodal transport (MT) on urban economic and environmental aspects. This paper integrates an optimal ridesharing matching method into a path-based deterministic day-to-day traffic assignment framework, considers match cancellations, and captures the interactions between various modes on the road. The model incorporates five traffic modes (solo driving, ridesharing as a driver, ridesharing as a passenger, bus travel, and metro travel) and two groups of travelers based on their ownership status. Its steady state is determined through numerical experiments. The sensitivity analyses reveal that the MT system's performance varies with changes in ownership, bus fare, and ridesharing fare, demonstrating diverse impacts on mode split, travel cost, and emissions across different groups, road links, and regions. Our findings suggest that vehicle restrictions and pricing strategies have both benefits and drawbacks in managing MT system, emphasizing the need for careful consideration of trade-offs and social equity implications in policy-making and implementation. This study not only enhances the theoretical understanding of MT system but also provides valuable support for urban transportation policy-making aimed at achieving efficient, sustainable, and socially equitable transport systems.

physics.soc-ph↗

arXiv2024

Using Multi-Source Data to Identify High-Emitting Heavy-Duty Diesel Vehicles

Identifying and managing high-emitters among heavy-duty diesel vehicles is a key to mitigating urban air pollution, as a small number of such vehicles could contribute a significant amount of total transport emissions. On-board monitoring (OBM) systems can directly monitor the real-time emission performance of heavy-duty vehicles on road and have become part of the future emissions compliance framework. The challenge, however, lies in the frequent unavailability of OBM data, affecting the effective screening of high-emitting vehicles. This work proposes to bridge the gap by integrating OBM data with remote sensing data to create a comprehensive monitoring system. OBM data is used to characterize the detailed real-world NOx emission performance of both normally-behaving vehicles and high-emitters at various vehicle operating conditions. Remote sensing data is employed to screen out candidate high-emitting vehicles based on thresholds determined by OBM data. Finally, the dynamic NOx emission reduction potential across all roads is mapped by combining the trajectory data for each vehicle with the emission data. A case study in Chengdu, China, utilizing emission and traffic data from heavy-duty vehicles for transporting construction waste (a.k.a. slag trucks), reveals the national threshold for identifying high-emitters via remote sensing might be too lenient, particularly in the medium speed range. An emission reduction of 18.8% in the China V slag truck fleet could be achieved by implementing this novel method in practice in Chengdu. This approach establishes a reliable and ongoing scheme for pinpointing high-emitters through multi-source data, which allows local authorities to develop more robust and targeted strategies to mitigate urban air pollution from heavy-duty diesel vehicles.

stat.AP↗

arXiv2023

OrthoPlanes: A Novel Representation for Better 3D-Awareness of GANs

We present a new method for generating realistic and view-consistent images with fine geometry from 2D image collections. Our method proposes a hybrid explicit-implicit representation called \textbf{OrthoPlanes}, which encodes fine-grained 3D information in feature maps that can be efficiently generated by modifying 2D StyleGANs. Compared to previous representations, our method has better scalability and expressiveness with clear and explicit information. As a result, our method can handle more challenging view-angles and synthesize articulated objects with high spatial degree of freedom. Experiments demonstrate that our method achieves state-of-the-art results on FFHQ and SHHQ datasets, both quantitatively and qualitatively. Project page: \url{https://orthoplanes.github.io/}.

cs.CV↗

arXiv2023

3DHumanGAN: 3D-Aware Human Image Generation with 3D Pose Mapping

We present 3DHumanGAN, a 3D-aware generative adversarial network that synthesizes photorealistic images of full-body humans with consistent appearances under different view-angles and body-poses. To tackle the representational and computational challenges in synthesizing the articulated structure of human bodies, we propose a novel generator architecture in which a 2D convolutional backbone is modulated by a 3D pose mapping network. The 3D pose mapping network is formulated as a renderable implicit function conditioned on a posed 3D human mesh. This design has several merits: i) it leverages the strength of 2D GANs to produce high-quality images; ii) it generates consistent images under varying view-angles and poses; iii) the model can incorporate the 3D human prior and enable pose conditioning. Project page: https://3dhumangan.github.io/.

cs.CV↗

arXiv2021

MoCaNet: Motion Retargeting in-the-wild via Canonicalization Networks

We present a novel framework that brings the 3D motion retargeting task from controlled environments to in-the-wild scenarios. In particular, our method is capable of retargeting body motion from a character in a 2D monocular video to a 3D character without using any motion capture system or 3D reconstruction procedure. It is designed to leverage massive online videos for unsupervised training, needless of 3D annotations or motion-body pairing information. The proposed method is built upon two novel canonicalization operations, structure canonicalization and view canonicalization. Trained with the canonicalization operations and the derived regularizations, our method learns to factorize a skeleton sequence into three independent semantic subspaces, i.e., motion, structure, and view angle. The disentangled representation enables motion retargeting from 2D to 3D with high precision. Our method achieves superior performance on motion transfer benchmarks with large body variations and challenging actions. Notably, the canonicalized skeleton sequence could serve as a disentangled and interpretable representation of human motion that benefits action analysis and motion retrieval.

cs.CV↗

arXiv2020

TransMoMo: Invariance-Driven Unsupervised Video Motion Retargeting

We present a lightweight video motion retargeting approach TransMoMo that is capable of transferring motion of a person in a source video realistically to another video of a target person. Without using any paired data for supervision, the proposed method can be trained in an unsupervised manner by exploiting invariance properties of three orthogonal factors of variation including motion, structure, and view-angle. Specifically, with loss functions carefully derived based on invariance, we train an auto-encoder to disentangle the latent representations of such factors given the source and target video clips. This allows us to selectively transfer motion extracted from the source video seamlessly to the target video in spite of structural and view-angle disparities between the source and the target. The relaxed assumption of paired data allows our method to be trained on a vast amount of videos needless of manual annotation of source-target pairing, leading to improved robustness against large structural variations and extreme motion in videos. We demonstrate the effectiveness of our method over the state-of-the-art methods. Code, model and data are publicly available on our project page (https://yzhq97.github.io/transmomo).

cs.CV↗

arXiv2019

Scene Graph Reasoning with Prior Visual Relationship for Visual Question Answering

One of the key issues of Visual Question Answering (VQA) is to reason with semantic clues in the visual content under the guidance of the question, how to model relational semantics still remains as a great challenge. To fully capture visual semantics, we propose to reason over a structured visual representation - scene graph, with embedded objects and inter-object relationships. This shows great benefit over vanilla vector representations and implicit visual relationship learning. Based on existing visual relationship models, we propose a visual relationship encoder that projects visual relationships into a learned deep semantic space constrained by visual context and language priors. Upon the constructed graph, we propose a Scene Graph Convolutional Network (SceneGCN) to jointly reason the object properties and relational semantics for the correct answer. We demonstrate the model's effectiveness and interpretability on the challenging GQA dataset and the classical VQA 2.0 dataset, remarkably achieving state-of-the-art 54.56% accuracy on GQA compared to the existing best model.

cs.MM↗

arXiv2019

Semantic Modeling of Textual Relationships in Cross-Modal Retrieval

Feature modeling of different modalities is a basic problem in current research of cross-modal information retrieval. Existing models typically project texts and images into one embedding space, in which semantically similar information will have a shorter distance. Semantic modeling of textural relationships is notoriously difficult. In this paper, we propose an approach to model texts using a featured graph by integrating multi-view textual relationships including semantic relations, statistical co-occurrence, and prior relations in the knowledge base. A dual-path neural network is adopted to learn multi-modal representations of information and cross-modal similarity measure jointly. We use a Graph Convolutional Network (GCN) for generating relation-aware text representations, and use a Convolutional Neural Network (CNN) with non-linearities for image representations. The cross-modal similarity measure is learned by distance metric learning. Experimental results show that, by leveraging the rich relational semantics in texts, our model can outperform the state-of-the-art models by 3.4% and 6.3% on accuracy on two benchmark datasets.

cs.MM↗