刘烨斌｜清华大学｜学者主页

2026

具身智能前沿展望：数据、模型与系统演进

穆尧，赵昊，胡瑞珍，张力，李弘扬，杨蛟龙，王靖博，韩磊，苏永峰，徐凯，杨易，李江，戴若犁，陈宝权，刘烨斌，弋力

《中国图象图形学报》

2026

2025年度三维视觉前沿趋势与十大进展

刘烨斌，穆尧，叶琦，高林，韩晓光，陈安沛，段岳圻，彭思达，邵天甲，张鸿文，张力，廖依伊，许岚，刘希慧，姚遥，胡瑞珍，弋力，郭裕兰，连宙辉，刘子纬，陈宝权

《中国图象图形学报》

2026

三维重建与生成前沿进展与展望

韩晓光，修宇亮，徐震，连宙辉，彭思达，姚遥，陈安沛，黄经纬，张邦，许岚，徐枫，章国锋，许威威，虞晶怡，刘利刚，陈宝权，刘烨斌，周晓巍

《中国图象图形学报》

2026

AI for VR:虚拟现实中的人工智能

王莉莉，许威威，刘烨斌，汪淼，王贝贝，杨旭波，许岚，谭张尧，樊润泽，王梓竣，王驰，张鸿文，温怡健，杨浩中，吴健，樊家辉，王晖，张启煊，汪国平，王涌天，赵沁平

《中国科学:信息科学》

2026

Erratum: OmniHands: Robust Motion Capture of Interactive Hands via A Versatile Transformer

Dixuan Lin, Yuxiang Zhang, Mengcheng Li, Jing Wei, Qi Yan, Qianying Wang, Yebin Liu, Hongwen Zhang

ACM Transactions on Graphics

精选作品

期刊论文

具身智能前沿展望：数据、模型与系统演进

2025年度三维视觉前沿趋势与十大进展

三维重建与生成前沿进展与展望

AI for VR:虚拟现实中的人工智能

Erratum: OmniHands: Robust Motion Capture of Interactive Hands via A Versatile Transformer

Mix3R: Mixing Feed-forward Reconstruction and Generative 3D Priors for Joint Multi-view Aligned 3D Reconstruction and Pose Estimation

OmniHands: Robust Motion Capture of Interactive Hands via A Versatile Transformer

Parametric Gaussian Human Model: Generalizable Prior for Efficient and Realistic Human Avatar Modeling

Tessellation GS: Neural Mesh Gaussians for Robust Monocular Reconstruction of Dynamic Objects

GAF: Gaussian Action Field as a 4D Representation for Dynamic World Modeling in Robotic Manipulation

ReCoM: Realistic Co-Speech Motion Generation with Recurrent Embedded Transformer

HOSIG: Full-Body Human-Object-Scene Interaction Generation with Hierarchical Scene Perception

Splat-SAP: Feed-Forward Gaussian Splatting for Human-Centered Scene with Scale-Aware Point Map Reconstruction

FOF-X: Towards Real-Time Detailed Human Reconstruction From a Single Image

HiReFF: High-Resolution Feedforward Human Reconstruction from Uncalibrated Sparse-View Video

HOSIG: Full-Body Human-Object-Scene Interaction Generation with Hierarchical Scene Perception

Splat-SAP: Feed-Forward Gaussian Splatting for Human-Centered Scene with Scale-Aware Point Map Reconstruction

场景重光照研究综述

2024年度三维视觉前沿趋势与十大进展

ManiVideo: Generating Hand-Object Manipulation Video with Dexterous and Generalizable Grasping

GBC-Splat: Generalizable Gaussian-Based Clothed Human Digitalization under Sparse RGB Cameras

EigenActor: Variant Body-Object Interaction Generation Evolved from Invariant Action Basis Reasoning

A Plug-And-Play Physical Motion Restoration Approach for In-The-Wild High-Difficulty Motions

SViMo: Synchronized Diffusion for Video and Motion Generation in Hand-object Interaction Scenarios

Fast and Globally Consistent Normal Orientation based on the Winding Number Normal Consistency

GPS-Gaussian: Generalizable Pixel-Wise 3D Gaussian Splatting for Real-Time Human Novel View Synthesis

Lodge: A Coarse to Fine Diffusion Network for Long Dance Generation Guided by the Characteristic Dance Primitives

HumanNorm: Learning Normal Diffusion Model for High-quality and Realistic 3D Human Generation

TACO: Benchmarking Generalizable Bimanual Tool-ACtion-Object Understanding

ProxyCap: Real-Time Monocular Full-Body Capture in World Space via Human-Centric Proxy-to-Motion Learning

TexVocab: Texture Vocabulary-Conditioned Human Avatars

MeshAvatar: Learning High-Quality Triangular Human Avatars from Multi-view Videos

3D Gaussian Parametric Head Model

MonoGaussianAvatar: Monocular Gaussian Point-based Head Avatar

LayGA: Layered Gaussian Avatars for Animatable Clothing Transfer

Tele-Aloha: A Telepresence System with Low-budget and High-authenticity Using Sparse RGB Cameras

InvertAvatar: Incremental GAN Inversion for Generalized Head Avatars

High-Fidelity 3D Head Avatars Reconstruction through Spatially-Varying Expression Conditioned Neural Radiance Field

Learning Explicit Contact for Implicit Reconstruction of Hand-Held Objects from Monocular Images

HAvatar: High-fidelity Head Avatar via Facial Model Conditioned Neural Radiance Field

AvatarReX: Real-time Expressive Full-body Avatars

Next3D: Generative Neural Texture Rasterization for 3D-Aware Head Avatars

CloSET: Modeling Clothed Humans on Continuous Surface with Explicit Template Decomposition

CaPhy: Capturing Physical Properties for Animatable Human Avatars

Narrator: Towards Natural Control of Human-Scene Interaction Generation via Relationship Reasoning

Leveraging Intrinsic Properties for Non-Rigid Garment Alignment

AvatarMAV: Fast 3D Head Avatar Reconstruction Using Motion-Aware Neural Voxels

StyleAvatar: Real-time Photo-realistic Portrait Avatar from a Single Video

LatentAvatar: Learning Latent Expression Code for Expressive Neural Head Avatar

PoseVocab: Learning Joint-structured Pose Embeddings for Human Avatar Modeling

Delving Deep into Pixel Alignment Feature for Accurate Multi-View Human Mesh Recovery

三维视觉与智能图形专刊简介

IDE-3D

HVTR: Hybrid Volumetric-Textural Rendering for Human Avatars

FENeRF: Face Editing in Neural Radiance Fields

Interacting Attention Graph for Single Image Two-Hand Reconstruction

DoubleField: Bridging the Neural Surface and Radiance Fields for High-fidelity Human Reconstruction and Rendering

High-Fidelity Human Avatars from a Single RGB Camera

DeepCloth: Neural Garment Representation for Shape and Style Editing

FloRen: Real-time High-quality Human Performance Rendering via Appearance Flow Using Sparse RGB Cameras

GIMO: Gaze-Informed Human Motion Prediction in Context

DiffuStereo: High Quality Human Reconstruction via Diffusion-Based Stereo Using Sparse Cameras

Learning Implicit Templates for Point-Based Clothed Human Modeling

Monocular Real-Time Human Geometry Reconstruction

三维视觉前沿进展

Cross-MPI: Cross-scale Stereo for Image Super-Resolution using Multiplane Images

Revisiting Light Field Rendering with Deep Anti-Aliasing Neural Network

Image-Guided Human Reconstruction via Multi-Scale Graph Transformation Networks

MulayCap: Multi-Layer Human Performance Capture Using a Monocular Video Camera

Model Study of Transient Imaging With Multi-Frequency Time-of-Flight Sensors

PoNA: Pose-Guided Non-Local Attention for Human Pose Transfer

NormalGAN: Learning Detailed 3D Human from a Single RGB-D Image

RobustFusion: Human Volumetric Capture with Data-Driven Visual Cues Using a RGBD Camera

SimulCap : Single-View Human Performance Capture With Cloth Simulation

Attention-guided GANs for human pose transfer

DeepHuman: 3D Human Reconstruction From a Single Image

3D Face Reprentation and Reconstruction with Multi-scale Graph Convolutional Autoencoders

基于瞬态成像技术的穿透散射介质成像