Qihang Yu

papers

1,240

total citations

papers (21)

Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIP

NEURIPS 2023arXiv

212

citations

Mask Guided Matting via Progressive Refinement Network

CVPR 2021arXiv

143

citations

C2FNAS: Coarse-to-Fine Neural Architecture Search for 3D Medical Image Segmentation

CVPR 2020arXiv

124

citations

Democratizing Text-to-Image Masked Generative Models with Compact Text-Aware One-Dimensional Tokens

ICCV 2025arXiv

citations

COCONut: Modernizing COCO Segmentation

CVPR 2024arXiv

citations

CancerUniT: Towards a Single Unified Model for Effective Detection, Segmentation, and Diagnosis of Eight Major Cancers Using a Large Collection of CT Scans

ICCV 2023arXiv

citations

ReMaX: Relaxing for Better Training on Efficient Panoptic Segmentation

NEURIPS 2023arXiv

citations

k-Means Mask Transformer

ECCV 2022arXiv

citations

FlowTok: Flowing Seamlessly Across Text and Image Tokens

ICCV 2025arXiv

citations

Compositor: Bottom-Up Clustering and Compositing for Robust Part and Object Segmentation

CVPR 2023arXiv

citations

Leveraging Panoptic Scene Graph for Evaluating Fine-Grained Text-to-Image Generation

ICCV 2025

citations

"PartImageNet: A Large, High-Quality Dataset of Parts"

ECCV 2022

citations

Qihang Yu

papers (21)

Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIP

Mask Guided Matting via Progressive Refinement Network

C2FNAS: Coarse-to-Fine Neural Architecture Search for 3D Medical Image Segmentation

CMT-DeepLab: Clustering Mask Transformers for Panoptic Segmentation

Randomized Autoregressive Visual Generation

Glance-and-Gaze Vision Transformer

MaskBit: Embedding-free Image Generation via Bit Tokens

Neural Architecture Search for Lightweight Non-Local Networks

Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generation

ViTamin: Designing Scalable Vision Models in the Vision-Language Era

TubeFormer-DeepLab: Video Mask Transformer

FlowAR: Scale-wise Autoregressive Image Generation Meets Flow Matching

Democratizing Text-to-Image Masked Generative Models with Compact Text-Aware One-Dimensional Tokens

COCONut: Modernizing COCO Segmentation

CancerUniT: Towards a Single Unified Model for Effective Detection, Segmentation, and Diagnosis of Eight Major Cancers Using a Large Collection of CT Scans

ReMaX: Relaxing for Better Training on Efficient Panoptic Segmentation

k-Means Mask Transformer

FlowTok: Flowing Seamlessly Across Text and Image Tokens

Compositor: Bottom-Up Clustering and Compositing for Robust Part and Object Segmentation

Leveraging Panoptic Scene Graph for Evaluating Fine-Grained Text-to-Image Generation

"PartImageNet: A Large, High-Quality Dataset of Parts"

papers (21)

Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIP

Mask Guided Matting via Progressive Refinement Network

C2FNAS: Coarse-to-Fine Neural Architecture Search for 3D Medical Image Segmentation

CMT-DeepLab: Clustering Mask Transformers for Panoptic Segmentation

Randomized Autoregressive Visual Generation

Glance-and-Gaze Vision Transformer

MaskBit: Embedding-free Image Generation via Bit Tokens

Neural Architecture Search for Lightweight Non-Local Networks

Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generation

ViTamin: Designing Scalable Vision Models in the Vision-Language Era

TubeFormer-DeepLab: Video Mask Transformer

FlowAR: Scale-wise Autoregressive Image Generation Meets Flow Matching

Democratizing Text-to-Image Masked Generative Models with Compact Text-Aware One-Dimensional Tokens

COCONut: Modernizing COCO Segmentation

CancerUniT: Towards a Single Unified Model for Effective Detection, Segmentation, and Diagnosis of Eight Major Cancers Using a Large Collection of CT Scans

ReMaX: Relaxing for Better Training on Efficient Panoptic Segmentation

k-Means Mask Transformer

FlowTok: Flowing Seamlessly Across Text and Image Tokens

Compositor: Bottom-Up Clustering and Compositing for Robust Part and Object Segmentation

Leveraging Panoptic Scene Graph for Evaluating Fine-Grained Text-to-Image Generation

"PartImageNet: A Large, High-Quality Dataset of Parts"