Jason Kuen

papers

941

total citations

papers (22)

DiffIP: Representation Fingerprints for Robust IP Protection of Diffusion Models

ICCV 2025

citations

MAGICK: A Large-scale Captioned Dataset from Matting Generated Images using Chroma Keying

CVPR 2024

citations

Jason Kuen

papers (22)

Multimodal Contrastive Training for Visual Representation Learning

SelfDoc: Self-Supervised Document Representation Learning

Open-Vocabulary Instance Segmentation via Robust Cross-Modal Pseudo-Labeling

High Quality Entity Segmentation

ImageFolder: Autoregressive Image Generation with Folded Tokens

Multi-Scale Aligned Distillation for Low-Resolution Detection

High Quality Segmentation for Ultra High-Resolution Images

SceneComposer: Any-Level Semantic Image Synthesis

LazyDiT: Lazy Learning for the Acceleration of Diffusion Transformers

Meta Spatio-Temporal Debiasing for Video Scene Graph Generation

TopNet: Transformer-Based Object Placement Network for Image Compositing

SegGen: Supercharging Segmentation Models with Text2Mask and Mask2Img Synthesis

CA-SSL: Class-Agnostic Semi-Supervised Learning for Detection and Segmentation

Improving the Reliability for Confidence Estimation

GALA: Toward Geometry-and-Lighting-Aware Object Search for Compositing

SOHES: Self-supervised Open-world Hierarchical Entity Segmentation

Refer to Any Segmentation Mask Group With Vision-Language Prompts

UniDoc: Unified Pretraining Framework for Document Understanding

AIMS: All-Inclusive Multi-Level Segmentation for Anything

Self-Supervised Relationship Probing

DiffIP: Representation Fingerprints for Robust IP Protection of Diffusion Models

MAGICK: A Large-scale Captioned Dataset from Matting Generated Images using Chroma Keying

papers (22)

Multimodal Contrastive Training for Visual Representation Learning

SelfDoc: Self-Supervised Document Representation Learning

Open-Vocabulary Instance Segmentation via Robust Cross-Modal Pseudo-Labeling

High Quality Entity Segmentation

ImageFolder: Autoregressive Image Generation with Folded Tokens

Multi-Scale Aligned Distillation for Low-Resolution Detection

High Quality Segmentation for Ultra High-Resolution Images

SceneComposer: Any-Level Semantic Image Synthesis

LazyDiT: Lazy Learning for the Acceleration of Diffusion Transformers

Meta Spatio-Temporal Debiasing for Video Scene Graph Generation

TopNet: Transformer-Based Object Placement Network for Image Compositing

SegGen: Supercharging Segmentation Models with Text2Mask and Mask2Img Synthesis

CA-SSL: Class-Agnostic Semi-Supervised Learning for Detection and Segmentation

Improving the Reliability for Confidence Estimation

GALA: Toward Geometry-and-Lighting-Aware Object Search for Compositing

SOHES: Self-supervised Open-world Hierarchical Entity Segmentation

Refer to Any Segmentation Mask Group With Vision-Language Prompts

UniDoc: Unified Pretraining Framework for Document Understanding

AIMS: All-Inclusive Multi-Level Segmentation for Anything

Self-Supervised Relationship Probing

DiffIP: Representation Fingerprints for Robust IP Protection of Diffusion Models

MAGICK: A Large-scale Captioned Dataset from Matting Generated Images using Chroma Keying