Pengxu Wei

papers

741

total citations

papers (16)

Component Divide-and-Conquer for Real-World Image Super-Resolution

ECCV 2020arXiv

324

citations

Identity-Preserving Talking Face Generation With Landmark and Appearance Priors

CVPR 2023arXiv

128

citations

Trash To Treasure: Harvesting OOD Data With Cross-Modal Matching for Open-Set Semi-Supervised Learning

ICCV 2021arXiv

citations

Towards Understanding the Robustness of Diffusion-Based Purification: A Stochastic Perspective

ICLR 2025arXiv

citations

Delving into Cascaded Instability: A Lipschitz Continuity View on Image Restoration and Object Detection Synergy

NEURIPS 2025arXiv

citations

CDNet: Centripetal Direction Network for Nuclear Instance Segmentation

ICCV 2021

citations

DigitalLLaVA: Incorporating Digital Cognition Capability for Physical World Comprehension in Multimodal LLMs

AAAI 2025

citations

TopoSeg: Topology-Aware Nuclear Instance Segmentation

ICCV 2023

citations

Kepler codebook

ICML 2024

citations

Can We Achieve Efficient Diffusion Without Self-Attention? Distilling Self-Attention into Convolutions

ICCV 2025arXiv

citations

Pengxu Wei

papers (16)

Component Divide-and-Conquer for Real-World Image Super-Resolution

Identity-Preserving Talking Face Generation With Landmark and Appearance Priors

Trash To Treasure: Harvesting OOD Data With Cross-Modal Matching for Open-Set Semi-Supervised Learning

Out-of-Candidate Rectification for Weakly Supervised Semantic Segmentation

Image Restoration Through Generalized Ornstein-Uhlenbeck Bridge

Adversarially-Aware Robust Object Detector

Towards Real-World Burst Image Super-Resolution: Benchmark and Method

Masked Images Are Counterfactual Samples for Robust Fine-Tuning

Dual Adversarial Adaptation for Cross-Device Real-World Image Super-Resolution

Towards Understanding the Robustness of Diffusion-Based Purification: A Stochastic Perspective

Delving into Cascaded Instability: A Lipschitz Continuity View on Image Restoration and Object Detection Synergy

CDNet: Centripetal Direction Network for Nuclear Instance Segmentation

DigitalLLaVA: Incorporating Digital Cognition Capability for Physical World Comprehension in Multimodal LLMs

TopoSeg: Topology-Aware Nuclear Instance Segmentation

Kepler codebook

Can We Achieve Efficient Diffusion Without Self-Attention? Distilling Self-Attention into Convolutions

papers (16)

Component Divide-and-Conquer for Real-World Image Super-Resolution

Identity-Preserving Talking Face Generation With Landmark and Appearance Priors

Trash To Treasure: Harvesting OOD Data With Cross-Modal Matching for Open-Set Semi-Supervised Learning

Out-of-Candidate Rectification for Weakly Supervised Semantic Segmentation

Image Restoration Through Generalized Ornstein-Uhlenbeck Bridge

Adversarially-Aware Robust Object Detector

Towards Real-World Burst Image Super-Resolution: Benchmark and Method

Masked Images Are Counterfactual Samples for Robust Fine-Tuning

Dual Adversarial Adaptation for Cross-Device Real-World Image Super-Resolution

Towards Understanding the Robustness of Diffusion-Based Purification: A Stochastic Perspective

Delving into Cascaded Instability: A Lipschitz Continuity View on Image Restoration and Object Detection Synergy

CDNet: Centripetal Direction Network for Nuclear Instance Segmentation

DigitalLLaVA: Incorporating Digital Cognition Capability for Physical World Comprehension in Multimodal LLMs

TopoSeg: Topology-Aware Nuclear Instance Segmentation

Kepler codebook

Can We Achieve Efficient Diffusion Without Self-Attention? Distilling Self-Attention into Convolutions