Fan Wang

48

Papers

118

Total Citations

1

Affiliations

Affiliations

Nanyang Technological University

Papers (48)

OmniSeg3D: Omniversal 3D Segmentation via Hierarchical Contrastive Learning

MVGenMaster: Scaling Multi-View Generation from Any Image via 3D Priors Enhanced Diffusion Model

AnimateAnyMesh: A Feed-Forward 4D Foundation Model for Text-Driven Universal Mesh Animation

FPSAttention: Training-Aware FP8 and Sparsity Co-Design for Fast Video Diffusion

RealisHuman: A Two-Stage Approach for Refining Malformed Human Parts in Generated Images

CoT-lized Diffusion: Let's Reinforce T2I Generation Step-by-step

PlayerOne: Egocentric World Simulator

Towards Large-Scale In-Context Reinforcement Learning by Meta-Training in Randomized Worlds

Distinguish Then Exploit: Source-free Open Set Domain Adaptation via Weight Barcode Estimation and Sparse Label Assignment

SparseDiT: Token Sparsification for Efficient Diffusion Transformer

Leveraging Imperfect Restoration for Data Availability Attack

RealisID: Scale-Robust and Fine-Controllable Identity Customization via Local and Global Complementation

Transfer Learning for Nonparametric Contextual Dynamic Pricing

Memory-Augmented Deep Conditional Unfolding Network for Pan-Sharpening

Decoupling and Recoupling Spatiotemporal Representation for RGB-D-Based Motion Recognition

Efficient Mask Correction for Click-Based Interactive Image Segmentation

A Practical Upper Bound for the Worst-Case Attribution Deviations

Foundation Model Drives Weakly Incremental Learning for Semantic Segmentation

Beyond Appearance: A Semantic Controllable Self-Supervised Learning Framework for Human-Centric Visual Tasks

MHPL: Minimum Happy Points Learning for Active Source Free Domain Adaptation

Making Vision Transformers Efficient From a Token Sparsification View

NIKI: Neural Inverse Kinematics With Invertible Neural Networks for 3D Human Pose and Shape Estimation

3D-Assisted Feature Synthesis for Novel Views of an Object

Probabilistic Deep Ordinal Regression Based on Gaussian Processes

TransReID: Transformer-Based Object Re-Identification

Revisiting Vision Transformer from the View of Path Ensemble

Dual Meta-Learning with Longitudinally Consistent Regularization for One-Shot Brain Tissue Segmentation Across the Human Lifespan

TopoGAN: A Topology-Aware Generative Adversarial Network

New Threats against Object Detector with Non-local Block

Unstructured Feature Decoupling for Vehicle Re-identification

KVT: k-NN Attention for Boosting Vision Transformers

TransFGU: A Top-down Approach to Fine-Grained Unsupervised Semantic Segmentation

Integrating Dashcam Views Through Inter-Video Mapping

Preacher: Paper-to-Video Agentic System

Enhancing Diffusion Model with Auxiliary Information Mining-Exploration and Efficient Sampling Mechanism for Sequential Recommendation

DR-VAE: Debiased and Representation-enhanced Variational Autoencoder for Collaborative Recommendation

BVT-IMA: Binary Vision Transformer with Information-Modified Attention

CL2CM: Improving Cross-Lingual Cross-Modal Retrieval via Cross-Lingual Knowledge Transfer

Exploring Channel-Aware Typical Features for Out-of-Distribution Detection

XScale-NVS: Cross-Scale Novel View Synthesis with Hash Featurized Manifold

DiffAug: Enhance Unsupervised Contrastive Learning with Domain-Knowledge-Free Diffusion-based Data Augmentation

Accelerating Parallel Sampling of Diffusion Models

Reducing Item Discrepancy via Differentially Private Robust Embedding Alignment for Privacy-Preserving Cross Domain Recommendation

Exploring Domain-Invariant Parameters for Source Free Domain Adaptation

EPro-PnP: Generalized End-to-End Probabilistic Perspective-N-Points for Monocular Object Pose Estimation

VTC-LFC: Vision Transformer Compression with Low-Frequency Components

Exploiting the Relationship Between Kendall's Rank Correlation and Cosine Similarity for Attribution Protection

Data Pruning via Moving-one-Sample-out