PaperHub
影响力指数
86.15/100
前 0.8%
全站排名 #542
发表论文24 篇
平均评分5.4
年均产出8.0 篇/年

Yuan Liu

Assistant Professor@Hong Kong University of Science and Technology·中国香港·OpenReview
研究方向

diffusion models for 3D tasks · neural representations and neural rendering · 6DoF object pose estimation · Point cloud registration · Image matching

5.5
19

PartSAM: A Scalable Promptable Part Segmentation Model Trained on Native 3D Data

ICLR 2026Poster
5.3
15

PAT3D: Physics-Augmented Text-to-3D Scene Generation

ICLR 2026Poster
4.5
15

SCoT: Teaching 3D-LLMs to Think Spatially with Million-scale CoT Annotations

ICLR 2026Poster
4.5
12

RefAny3D: 3D Asset-Referenced Diffusion Models for Image Generation

ICLR 2026Poster
4.5
6

PanoLAM: Large Avatar Model for Gaussian Full-Head Synthesis from One-shot Unposed Image

ICLR 2026Rejected
4.5
13

Jigsaw3D: Disentangled 3D Style Transfer via Patch Shuffling and Masking

ICLR 2026Rejected
通讯
4.4
23

MEGS^{2}: Memory-Efficient Gaussian Splatting via Spherical Gaussians and Unified Pruning

ICLR 2026Poster
4.0
16

TRELLISWorld: Training-Free World Generation from Object Generators

ICLR 2026Rejected
二作
4.0
4

MeshMosaic: Scaling Artist Mesh Generation via Local-to-Global Assembly

ICLR 2026Withdrawn
3.5
7

Prior-based 4D Human-Scene Reconstruction from Monocular Videos

ICLR 2026Withdrawn
3.0
5

T-MLP: Tailed Multi-Layer Perceptron for Level-of-Detail Signal Representation

ICLR 2026Withdrawn
8.2
22

🎧MOSPA: Human Motion Generation Driven by Spatial Audio

NeurIPS 2025Spotlight
7.3
26

TrackingWorld: World-centric Monocular 3D Tracking of Almost All Pixels

NeurIPS 2025Poster
通讯
6.8
30

PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation

NeurIPS 2025Poster
6.8
28

MoDGS: Dynamic Gaussian Splatting from Casually-captured Monocular Videos with Depth Priors

ICLR 2025Poster
二作
6.5
26

CityAnchor: City-scale 3D Visual Grounding with Multi-modality LLMs

ICLR 2025Poster
6.0
31

SyncHuman: Synchronizing 2D and 3D Generative Models for Single-view Human Reconstruction

NeurIPS 2025Poster
通讯
6.0
26

MVTokenFlow: High-quality 4D Content Generation using Multiview Token Flow

ICLR 2025Poster
二作
6.0
17

DICE: End-to-end Deformation Capture of Hand-Face Interactions from a Single Image

ICLR 2025Poster
5.0
5

PSHuman: Photorealistic Single-view Human Reconstruction using Cross-Scale Diffusion

ICLR 2025Withdrawn
三作
3.7
4

PFAvatar: Avatar Reconstruction from Multiple In-the-wild Images

ICLR 2025Withdrawn
二作
合作者 (20)