arXiv subjects

Yunfei Liu

Publications and source records attributed to Yunfei Liu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Astra: Toward General-Purpose Mobile Robots via Hierarchical Multimodal Learning

Modern robot navigation systems encounter difficulties in diverse and complex indoor environments. Traditional approaches rely on multiple modules with small models or rule-based systems and thus lack adaptability to new environments. To address this, we developed Astra, a comprehensive dual-model architecture, Astra-Global and Astra-Local, for mobile robot navigation. Astra-Global, a multimodal LLM, processes vision and language inputs to perform self and goal localization using a hybrid topological-semantic graph as the global map, and outperforms traditional visual place recognition methods. Astra-Local, a multitask network, handles local path planning and odometry estimation. Its 4D spatial-temporal encoder, trained through self-supervised learning, generates robust 4D features for downstream tasks. The planning head utilizes flow matching and a novel masked ESDF loss to minimize collision risks for generating local trajectories, and the odometry head integrates multi-sensor inputs via a transformer encoder to predict the relative pose of the robot. Deployed on real in-house mobile robots, Astra achieves high end-to-end mission success rate across diverse indoor environments.

cs.RO↗

arXiv2025

HRAvatar: High-Quality and Relightable Gaussian Head Avatar

Reconstructing animatable and high-quality 3D head avatars from monocular videos, especially with realistic relighting, is a valuable task. However, the limited information from single-view input, combined with the complex head poses and facial movements, makes this challenging. Previous methods achieve real-time performance by combining 3D Gaussian Splatting with a parametric head model, but the resulting head quality suffers from inaccurate face tracking and limited expressiveness of the deformation model. These methods also fail to produce realistic effects under novel lighting conditions. To address these issues, we propose HRAvatar, a 3DGS-based method that reconstructs high-fidelity, relightable 3D head avatars. HRAvatar reduces tracking errors through end-to-end optimization and better captures individual facial deformations using learnable blendshapes and learnable linear blend skinning. Additionally, it decomposes head appearance into several physical properties and incorporates physically-based shading to account for environmental lighting. Extensive experiments demonstrate that HRAvatar not only reconstructs superior-quality heads but also achieves realistic visual effects under varying lighting conditions.

cs.CV↗

arXiv2025

TEASER: Token Enhanced Spatial Modeling for Expressions Reconstruction

3D facial reconstruction from a single in-the-wild image is a crucial task in human-centered computer vision tasks. While existing methods can recover accurate facial shapes, there remains significant space for improvement in fine-grained expression capture. Current approaches struggle with irregular mouth shapes, exaggerated expressions, and asymmetrical facial movements. We present TEASER (Token EnhAnced Spatial modeling for Expressions Reconstruction), which addresses these challenges and enhances 3D facial geometry performance. TEASER tackles two main limitations of existing methods: insufficient photometric loss for self-reconstruction and inaccurate localization of subtle expressions. We introduce a multi-scale tokenizer to extract facial appearance information. Combined with a neural renderer, these tokens provide precise geometric guidance for expression reconstruction. Furthermore, TEASER incorporates a pose-dependent landmark loss to further improve geometric performances. Our approach not only significantly enhances expression reconstruction quality but also offers interpretable tokens suitable for various downstream applications, such as photorealistic facial video driving, expression transfer, and identity swapping. Quantitative and qualitative experimental results across multiple datasets demonstrate that TEASER achieves state-of-the-art performance in precise expression reconstruction.

cs.CV↗

arXiv2025

Exploring Iterative Manifold Constraint for Zero-shot Image Editing

Editability and fidelity are two essential demands for text-driven image editing, which expects that the editing area should align with the target prompt and the rest remain unchanged separately. The current cutting-edge editing methods usually obey an "inversion-then-editing" pipeline, where the input image is inverted to an approximate Gaussian noise ${z}_T$, based on which a sampling process is conducted using the target prompt. Nevertheless, we argue that it is not a good choice to use a near-Gaussian noise as a pivot for further editing since it would bring plentiful fidelity errors. We verify this by a pilot analysis, discovering that intermediate-inverted latents can achieve a better trade-off between editability and fidelity than the fully-inverted ${z}_T$. Based on this, we propose a novel zero-shot editing paradigm dubbed ZZEdit, which first locates a qualified intermediate-inverted latent marked as ${z}_p$ as a better editing pivot, which is sufficient-for-editing while structure-preserving. Then, a ZigZag process is designed to execute denoising and inversion alternately, which progressively inject target guidance to ${z}_p$ while preserving the structure information of $p$ step. Afterwards, to achieve the same step number of inversion and denoising, we execute a pure sampling process under the target prompt. Essentially, our ZZEdit performs iterative manifold constraint between the manifold of $M_{p}$ and $M_{p-1}$, leading to fewer fidelity errors. Extensive experiments highlight the effectiveness of ZZEdit in diverse image editing scenarios compared with the "inversion-then-editing" pipeline.

cs.CV↗

arXiv2025

Identity-Preserving Video Dubbing Using Motion Warping

Video dubbing aims to synthesize realistic, lip-synced videos from a reference video and a driving audio signal. Although existing methods can accurately generate mouth shapes driven by audio, they often fail to preserve identity-specific features, largely because they do not effectively capture the nuanced interplay between audio cues and the visual attributes of reference identity . As a result, the generated outputs frequently lack fidelity in reproducing the unique textural and structural details of the reference identity. To address these limitations, we propose IPTalker, a novel and robust framework for video dubbing that achieves seamless alignment between driving audio and reference identity while ensuring both lip-sync accuracy and high-fidelity identity preservation. At the core of IPTalker is a transformer-based alignment mechanism designed to dynamically capture and model the correspondence between audio features and reference images, thereby enabling precise, identity-aware audio-visual integration. Building on this alignment, a motion warping strategy further refines the results by spatially deforming reference images to match the target audio-driven configuration. A dedicated refinement process then mitigates occlusion artifacts and enhances the preservation of fine-grained textures, such as mouth details and skin features. Extensive qualitative and quantitative evaluations demonstrate that IPTalker consistently outperforms existing approaches in terms of realism, lip synchronization, and identity retention, establishing a new state of the art for high-quality, identity-consistent video dubbing.

cs.CV↗

arXiv2024

Constrained Online Recursive Source Separation Framework for Real-time Electrophysiological Signal Processing

Background and Objective: Processing electrophysiological signals often requires blind source separation (BSS) due to the nature of mixing source signals. However, its complex computational demands make real-time BSS challenging. The objective of this work is to develop an advanced real-time BSS method suitable for processing electrophysiological signals. Methods: In this paper, a novel BSS framework termed constrained online recursive source separation (CORSS) was proposed. In the framework, a stepwise recursive unmixing matrix learning rule was adopted to enable real-time updates with minimal computational cost. Moreover, by incorporating prior information of target signals to optimize the cost function, the framework algorithm was more likely to converge to the target sources. To validate its performance, the proposed framework was applied to both downstream tasks, namely real-time surface electromyogram (sEMG) decomposition and real-time respiratory intent monitoring based on diaphragmatic electromyogram (sEMGdi) extraction. Results: The proposed method achieved a matching rate of 96.00 % for the sEMG decomposition task and 98.12 % for the sEMGdi extraction task, exhibiting superior performance over other comparison methods (p < 0.05). Our method also exhibited minimal time delay during computation, with only 12.5 ms delay when the block size was 0.1s, demonstrating its good capabilities in online processing. Conclusions: The proposed method was demonstrated to enable real-time BSS with both improved separation performance and low computational latency. It is of substantial importance for real-time electrophysiological signal processing and applications towards advanced neural-machine interaction and clinical monitoring.

cs.HC↗

arXiv2024

DreamHead: Learning Spatial-Temporal Correspondence via Hierarchical Diffusion for Audio-driven Talking Head Synthesis

Audio-driven talking head synthesis strives to generate lifelike video portraits from provided audio. The diffusion model, recognized for its superior quality and robust generalization, has been explored for this task. However, establishing a robust correspondence between temporal audio cues and corresponding spatial facial expressions with diffusion models remains a significant challenge in talking head generation. To bridge this gap, we present DreamHead, a hierarchical diffusion framework that learns spatial-temporal correspondences in talking head synthesis without compromising the model's intrinsic quality and adaptability.~DreamHead learns to predict dense facial landmarks from audios as intermediate signals to model the spatial and temporal correspondences.~Specifically, a first hierarchy of audio-to-landmark diffusion is first designed to predict temporally smooth and accurate landmark sequences given audio sequence signals. Then, a second hierarchy of landmark-to-image diffusion is further proposed to produce spatially consistent facial portrait videos, by modeling spatial correspondences between the dense facial landmark and appearance. Extensive experiments show that proposed DreamHead can effectively learn spatial-temporal consistency with the designed hierarchical diffusion and produce high-fidelity audio-driven talking head videos for multiple identities.

cs.MM↗

arXiv2024

Temperature Secret in Bathtub: A Model of Temperature Distribution of Bathtub Based on Heat Conduction Equation

We use the multidimensional heat conduction and heat transfer equations to model the temperature distribution of water in a bathtub by solving partial differential equations. We address optimal water addition and bathtub design. First, we establish a water surface cooling model using Newton's law of cooling to simulate heat exchange between air and water. Without new heat sources, the water temperature reaches a minimum in 40 minutes. We then simulate adding hot water with a one-dimensional heat conduction model, including air cooling effects. We determine that the optimal heat input is 80 Joules and the optimal water velocity is 0.042 m/s to maintain temperature and save water. The ideal bathtub dimensions are 1.5m length, 0.6m width, 0.42m depth, with rounded corners. Using finite difference methods and MATLAB's Pdetool, we solve the heat conduction equation and verify numerical stability, discussing the model's pros and cons and suggesting improvements.

eess.SY↗

arXiv2024

Human Leg Training Machine Based on The Multi-linkage System

In real life, many people have leg defects. the goal of our work is to design a mechanism which could help them walk based on a specific trajectory and realize flexible walking finally. In this paper, we use a motor to drive a multi-link leg mechanism. The major issues addressed in this paper are as follows: (i) design human leg training mechanism based on the multi-link mechanism (ii) Simulate leg movement trajectory of multi-link mechanism based on walking process (iii) make use of one motor torque control to control the trajectory and velocity of this mechanism.

eess.SY↗

arXiv2024

Gravity Balanced Arm Exoskeleton for Basketball Shooting Training

This paper proposes a gravity balanced arm exoskeleton design for basketball shooting training. The potential energy equation of the mechanism is derived. A simulation of the arm going through the basketball shooting motion is done on the mechanism. Throughout the motion the total potential energy is constant. Thus, the proposed arm exoskeleton is indeed gravity balanced with the use of two springs.

eess.SY↗

arXiv2024

Solve the Refugee Crisis with Data

In this study, we addressed the refugee crisis through two main models. For predicting the ultimate number of refugees, we first established a Logistic Regression Model, but due to the limited data points, its prediction accuracy was suboptimal. Consequently, we incorporated Gray Theory to develop the Gary Verhulst Model, which provided scientifically sound and reasonable predictions. Statistical tests comparing both models highlighted the superiority of the Gary Verhulst Model. For formulating refugee allocation schemes, we initially used the Factor Analysis Method but found it too subjective and lacking in rigorous validation measures. We then developed a Refugee Allocation Model based on the Analytic Hierarchy Process (AHP), which absorbed the advantages of the former method. This model underwent extensive validation and passed consistency checks, resulting in an effective and scientific refugee allocation scheme. We also compared our model with the current allocation schemes and proposed improvements. Finally, we discussed the advantages and disadvantages of our models, their applicability, and scalability. Sensitivity analysis was conducted, and directions for future model improvements were identified.

cs.CY↗

arXiv2024

Improvement of Sensitivity of Capacitive Micromachined Ultrasound Transducer

Capacitive Micromachined Ultrasonic Transducer (CMUT) has a wild range of applications in medical detecting and imaging fields. However, operating under self-generating-self-receiving (SGSR) method usually results in poor sensitivity. But the sensitivity cannot be improved simply by increasing the resonant frequency since the frequency of a specific kind of CMUT is designed for specific usage. In this paper, based on one specific type of CMUT, mechanical model is built and simulation analysis is demonstrated. A brand-new method one-generating-multiple-receiving (OGMR) is introduced and a special circuit model has been designed to improve the signal to thermal noise ratio. By increasing the number of receiving capacitors from 1 to 8, we increased the signal-noise ratio to 2.83 times.

eess.SP↗

arXiv2024

Revisiting Modularity Maximization for Graph Clustering: A Contrastive Learning Perspective

Graph clustering, a fundamental and challenging task in graph mining, aims to classify nodes in a graph into several disjoint clusters. In recent years, graph contrastive learning (GCL) has emerged as a dominant line of research in graph clustering and advances the new state-of-the-art. However, GCL-based methods heavily rely on graph augmentations and contrastive schemes, which may potentially introduce challenges such as semantic drift and scalability issues. Another promising line of research involves the adoption of modularity maximization, a popular and effective measure for community detection, as the guiding principle for clustering tasks. Despite the recent progress, the underlying mechanism of modularity maximization is still not well understood. In this work, we dig into the hidden success of modularity maximization for graph clustering. Our analysis reveals the strong connections between modularity maximization and graph contrastive learning, where positive and negative examples are naturally defined by modularity. In light of our results, we propose a community-aware graph clustering framework, coined MAGI, which leverages modularity maximization as a contrastive pretext task to effectively uncover the underlying information of communities in graphs, while avoiding the problem of semantic drift. Extensive experiments on multiple graph datasets verify the effectiveness of MAGI in terms of scalability and clustering performance compared to state-of-the-art graph clustering methods. Notably, MAGI easily scales a sufficiently large graph with 100M nodes while outperforming strong baselines.

cs.LG↗

arXiv2024

A Video is Worth 256 Bases: Spatial-Temporal Expectation-Maximization Inversion for Zero-Shot Video Editing

This paper presents a video inversion approach for zero-shot video editing, which models the input video with low-rank representation during the inversion process. The existing video editing methods usually apply the typical 2D DDIM inversion or naive spatial-temporal DDIM inversion before editing, which leverages time-varying representation for each frame to derive noisy latent. Unlike most existing approaches, we propose a Spatial-Temporal Expectation-Maximization (STEM) inversion, which formulates the dense video feature under an expectation-maximization manner and iteratively estimates a more compact basis set to represent the whole video. Each frame applies the fixed and global representation for inversion, which is more friendly for temporal consistency during reconstruction and editing. Extensive qualitative and quantitative experiments demonstrate that our STEM inversion can achieve consistent improvement on two state-of-the-art video editing methods. Project page: https://stem-inv.github.io/page/.

cs.CV↗

arXiv2024

Revealing Hierarchical Structure of Leaf Venations in Plant Science via Label-Efficient Segmentation: Dataset and Method

Hierarchical leaf vein segmentation is a crucial but under-explored task in agricultural sciences, where analysis of the hierarchical structure of plant leaf venation can contribute to plant breeding. While current segmentation techniques rely on data-driven models, there is no publicly available dataset specifically designed for hierarchical leaf vein segmentation. To address this gap, we introduce the HierArchical Leaf Vein Segmentation (HALVS) dataset, the first public hierarchical leaf vein segmentation dataset. HALVS comprises 5,057 real-scanned high-resolution leaf images collected from three plant species: soybean, sweet cherry, and London planetree. It also includes human-annotated ground truth for three orders of leaf veins, with a total labeling effort of 83.8 person-days. Based on HALVS, we further develop a label-efficient learning paradigm that leverages partial label information, i.e. missing annotations for tertiary veins. Empirical studies are performed on HALVS, revealing new observations, challenges, and research directions on leaf vein segmentation.

cs.CV↗

arXiv2024

DiffSHEG: A Diffusion-Based Approach for Real-Time Speech-driven Holistic 3D Expression and Gesture Generation

We propose DiffSHEG, a Diffusion-based approach for Speech-driven Holistic 3D Expression and Gesture generation with arbitrary length. While previous works focused on co-speech gesture or expression generation individually, the joint generation of synchronized expressions and gestures remains barely explored. To address this, our diffusion-based co-speech motion generation transformer enables uni-directional information flow from expression to gesture, facilitating improved matching of joint expression-gesture distributions. Furthermore, we introduce an outpainting-based sampling strategy for arbitrary long sequence generation in diffusion models, offering flexibility and computational efficiency. Our method provides a practical solution that produces high-quality synchronized expression and gesture generation driven by speech. Evaluated on two public datasets, our approach achieves state-of-the-art performance both quantitatively and qualitatively. Additionally, a user study confirms the superiority of DiffSHEG over prior approaches. By enabling the real-time generation of expressive and synchronized motions, DiffSHEG showcases its potential for various applications in the development of digital humans and embodied agents.

cs.SD↗

arXiv2024

GPAvatar: Generalizable and Precise Head Avatar from Image(s)

Head avatar reconstruction, crucial for applications in virtual reality, online meetings, gaming, and film industries, has garnered substantial attention within the computer vision community. The fundamental objective of this field is to faithfully recreate the head avatar and precisely control expressions and postures. Existing methods, categorized into 2D-based warping, mesh-based, and neural rendering approaches, present challenges in maintaining multi-view consistency, incorporating non-facial information, and generalizing to new identities. In this paper, we propose a framework named GPAvatar that reconstructs 3D head avatars from one or several images in a single forward pass. The key idea of this work is to introduce a dynamic point-based expression field driven by a point cloud to precisely and effectively capture expressions. Furthermore, we use a Multi Tri-planes Attention (MTA) fusion module in the tri-planes canonical field to leverage information from multiple input images. The proposed method achieves faithful identity reconstruction, precise expression control, and multi-view consistency, demonstrating promising results for free-viewpoint rendering and novel view synthesis.

cs.CV↗

arXiv2023

LasTGL: An Industrial Framework for Large-Scale Temporal Graph Learning

Over the past few years, graph neural networks (GNNs) have become powerful and practical tools for learning on (static) graph-structure data. However, many real-world applications, such as social networks and e-commerce, involve temporal graphs where nodes and edges are dynamically evolving. Temporal graph neural networks (TGNNs) have progressively emerged as an extension of GNNs to address time-evolving graphs and have gradually become a trending research topic in both academics and industry. Advancing research and application in such an emerging field necessitates the development of new tools to compose TGNN models and unify their different schemes for dealing with temporal graphs. In this work, we introduce LasTGL, an industrial framework that integrates unified and extensible implementations of common temporal graph learning algorithms for various advanced tasks. The purpose of LasTGL is to provide the essential building blocks for solving temporal graph learning tasks, focusing on the guiding principles of user-friendliness and quick prototyping on which PyTorch is based. In particular, LasTGL provides comprehensive temporal graph datasets, TGNN models and utilities along with well-documented tutorials, making it suitable for both absolute beginners and expert deep learning practitioners alike.

cs.LG↗