Zirui Zhou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zirui Zhou has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Junfeng Ni
Tsinghua University, Beijing, China
1 shared publication - Liang Yue
Department of Computer Science and Technology, Tsinghua University, Beijing, China
1 shared publication - Lin Yuan
Tsinghua University, Beijing, China
1 shared publication - Nan Jiang
Institute for AI, Peking University, Beijing, China and National Key Lab of General AI, BIGAI, Beijing, China
1 shared publication - Shao-Kui Zhang
Beijing Normal University, Beijing, China
1 shared publication - Siyuan Huang
Beijing Institute for General Artificial Intelligence (BIGAI)
1 shared publication
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (6) , 1-12
DecomVoxel: Harnessing 3D-Native Priors with Guided In-situ Denoising Optimization for Decompositional Scene Reconstruction
Decompositional scene reconstruction aims to reconstruct high-quality objects and background, yet existing methods still struggle with the level of quality under heavy occlusions. While generative priors offer a potential solution, 2D image-based priors often suffer from multi-view inconsistency due to a lack of 3D awareness. Conversely, 3D-native priors provide stronger structural inductive biases but frequently lead to spatial drift and misalignment within complex scenes. To address these issues, we propose DecomVoxel, formulating object completion as a guided in-situ denoising optimization that bridges 3D-native priors with neural scene reconstruction. Our framework introduces a reformulated epsilon-based distillation loss to ensure stable latent refinement, alongside adaptive spatial guidance that utilizes occupied and vacant anchors with temporal annealing to suppress generative hallucinations and mitigate spatial drift. Experiments on Replica and ScanNet++ show that DecomVoxel significantly outperforms state-of-the-art methods while faithfully preserving the original spatial layout, structural fidelity, and style-consistent texture. Our method pushes the boundary of decompositional reconstruction by delivering high-quality textured meshes with clean topology, geometry, and appearance, providing a robust solution for the decompositional reconstruction of complex real-world scenes. Code is available at https://github.com/DecomVoxel/DecomVoxel.
- 9 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (6) , 1-12
Open AccessDecomVoxel: Harnessing 3D-Native Priors with Guided In-situ Denoising Optimization for Decompositional Scene Reconstruction
Decompositional scene reconstruction aims to reconstruct high-quality objects and background, yet existing methods still struggle with the level of quality under heavy occlusions. While generative priors offer a potential solution, 2D image-based priors often suffer from multi-view inconsistency due to a lack of 3D awareness. Conversely, 3D-native priors provide stronger structural inductive biases but frequently lead to spatial drift and misalignment within complex scenes. To address these issues, we propose DecomVoxel, formulating object completion as a guided in-situ denoising optimization that bridges 3D-native priors with neural scene reconstruction. Our framework introduces a reformulated epsilon-based distillation loss to ensure stable latent refinement, alongside adaptive spatial guidance that utilizes occupied and vacant anchors with temporal annealing to suppress generative hallucinations and mitigate spatial drift. Experiments on Replica and ScanNet++ show that DecomVoxel significantly outperforms state-of-the-art methods while faithfully preserving the original spatial layout, structural fidelity, and style-consistent texture. Our method pushes the boundary of decompositional reconstruction by delivering high-quality textured meshes with clean topology, geometry, and appearance, providing a robust solution for the decompositional reconstruction of complex real-world scenes. Code is available at https://github.com/DecomVoxel/DecomVoxel.
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 9385-9394
Synthesizing 3D Scenes via Diffusion Model that Incorporates Indoor Scene Characteristics
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 9385-9394
Open AccessSynthesizing 3D Scenes via Diffusion Model that Incorporates Indoor Scene Characteristics