Cheng Sun

Affiliations

NVIDIA

papers

1,673

total citations

papers (15)

Direct Voxel Grid Optimization: Super-Fast Convergence for Radiance Fields Reconstruction

CVPR 2022arXiv

1,283

citations

HoHoNet: 360 Indoor Holistic Understanding With Latent Horizontal Features

CVPR 2021arXiv

154

citations

SAM4MLLM: Enhance Multi-Modal Large Language Model for Referring Expression Segmentation

ECCV 2024arXiv

citations

Neural-PBIR Reconstruction of Shape, Material, and Illumination

ICCV 2023arXiv

citations

Sparse Voxels Rasterization: Real-time High-fidelity Radiance Field Rendering

CVPR 2025arXiv

citations

FrugalNeRF: Fast Convergence for Extreme Few-shot Novel View Synthesis without Learned Priors

CVPR 2025arXiv

citations

ImGeoNet: Image-induced Geometry-aware Voxel Representation for Multi-view 3D Object Detection

ICCV 2023arXiv

citations

Specialize and Fuse: Pyramidal Output Representation for Semantic Segmentation

ICCV 2021arXiv

citations

Hashing Neural Video Decomposition with Multiplicative Residuals in Space-Time

ICCV 2023arXiv

citations

Seg2Reg: Differentiable 2D Segmentation to 1D Regression Rendering for 360 Room Layout Reconstruction

CVPR 2024arXiv

citations

Multiview Regenerative Morphing with Dual Flows

ECCV 2022arXiv

citations

Segment Anything, Even Occluded

CVPR 2025arXiv

citations

Cheng Sun

Affiliations

papers (15)

Direct Voxel Grid Optimization: Super-Fast Convergence for Radiance Fields Reconstruction

HoHoNet: 360 Indoor Holistic Understanding With Latent Horizontal Features

SAM4MLLM: Enhance Multi-Modal Large Language Model for Referring Expression Segmentation

Neural-PBIR Reconstruction of Shape, Material, and Illumination

Sparse Voxels Rasterization: Real-time High-fidelity Radiance Field Rendering

FrugalNeRF: Fast Convergence for Extreme Few-shot Novel View Synthesis without Learned Priors

Indoor Panorama Planar 3D Reconstruction via Divide and Conquer

Data Efficient 3D Learner via Knowledge Transferred from 2D Model

LongSplat: Robust Unposed 3D Gaussian Splatting for Casual Long Videos

ImGeoNet: Image-induced Geometry-aware Voxel Representation for Multi-view 3D Object Detection

Specialize and Fuse: Pyramidal Output Representation for Semantic Segmentation

Hashing Neural Video Decomposition with Multiplicative Residuals in Space-Time

Seg2Reg: Differentiable 2D Segmentation to 1D Regression Rendering for 360 Room Layout Reconstruction

Multiview Regenerative Morphing with Dual Flows

Segment Anything, Even Occluded

papers (15)

Direct Voxel Grid Optimization: Super-Fast Convergence for Radiance Fields Reconstruction

HoHoNet: 360 Indoor Holistic Understanding With Latent Horizontal Features

SAM4MLLM: Enhance Multi-Modal Large Language Model for Referring Expression Segmentation

Neural-PBIR Reconstruction of Shape, Material, and Illumination

Sparse Voxels Rasterization: Real-time High-fidelity Radiance Field Rendering

FrugalNeRF: Fast Convergence for Extreme Few-shot Novel View Synthesis without Learned Priors

Indoor Panorama Planar 3D Reconstruction via Divide and Conquer

Data Efficient 3D Learner via Knowledge Transferred from 2D Model

LongSplat: Robust Unposed 3D Gaussian Splatting for Casual Long Videos

ImGeoNet: Image-induced Geometry-aware Voxel Representation for Multi-view 3D Object Detection

Specialize and Fuse: Pyramidal Output Representation for Semantic Segmentation

Hashing Neural Video Decomposition with Multiplicative Residuals in Space-Time

Seg2Reg: Differentiable 2D Segmentation to 1D Regression Rendering for 360 Room Layout Reconstruction

Multiview Regenerative Morphing with Dual Flows

Segment Anything, Even Occluded