Hao-Xiang Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Hao-Xiang Chen has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Tai-Jiang Mu
Tsinghua University
8 shared publications - Shi-Min Hu
Tsinghua University
5 shared publications - Qi-Yuan Feng
CS Dept, Tsinghua University, BEIJING, China
3 shared publications - Qun-Ce Xu
BNRist, Department of Computer Science and Technology, Tsinghua University, Beijing, China
3 shared publications - Jun-Xiong Cai
Tsinghua University Tsinghua University Tsinghua University Tsinghua University Cardiff University Tsinghua University
2 shared publications - Xiao-Lei Li
Tsinghua University, Beijing, China and Tencent Video AI Center, Beijing, China
2 shared publications
Latest Publications
- Open Access
- 11 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 46 (1) , 1-18
Grow3D: Hierarchical Next-Scale Octree Prediction for Fast and High-Fidelity 3D Shape Generation
Diffusion models and auto-regressive models have been widely adopted in recent research on 3D shape generation. However, diffusion models and standard next-token prediction auto-regressive models require dozens to even hundreds of generation steps, leading to slow inference. A new paradigm of auto-regressive with next-scale prediction has demonstrated both inference efficiency and generation quality. However, existing works are restricted to fixed-length representations and cannot be directly applied to more adaptive variable-length representations, such as sparse voxels. To address these limitations, we propose Grow3D , a new auto-regressive generative framework that generates high-quality 3D shapes in a coarse-to-fine manner via ”next-scale prediction”, achieving both high quality and fast inference. Specifically, we first employ a Vector Quantized Variational Autoencoder (VQ-VAE) with residual serialization to encode 3D shapes into a multiscale, sparse-structured latent representation with quantized features. Building on this latent representation, we utilize a next-scale prediction strategy to auto-regressively generate both the octree structure and the corresponding geometry features. Benefiting from the inherent structure of the octree, we introduce an octree-structure-aware attention mechanism that selectively attends to the most relevant features. Furthermore, a CFG-based sampling strategy is proposed to enhance the quality and diversity of generation. Extensive experiments demonstrate that Grow3D outperforms state-of-the-art methods in both 3D shape generation quality and speed, enabling some real-time downstream applications, such as interactive 3D editing.
- 11 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 46 (1) , 1-18
Open AccessGrow3D: Hierarchical Next-Scale Octree Prediction for Fast and High-Fidelity 3D Shape Generation
Diffusion models and auto-regressive models have been widely adopted in recent research on 3D shape generation. However, diffusion models and standard next-token prediction auto-regressive models require dozens to even hundreds of generation steps, leading to slow inference. A new paradigm of auto-regressive with next-scale prediction has demonstrated both inference efficiency and generation quality. However, existing works are restricted to fixed-length representations and cannot be directly applied to more adaptive variable-length representations, such as sparse voxels. To address these limitations, we propose Grow3D , a new auto-regressive generative framework that generates high-quality 3D shapes in a coarse-to-fine manner via ”next-scale prediction”, achieving both high quality and fast inference. Specifically, we first employ a Vector Quantized Variational Autoencoder (VQ-VAE) with residual serialization to encode 3D shapes into a multiscale, sparse-structured latent representation with quantized features. Building on this latent representation, we utilize a next-scale prediction strategy to auto-regressively generate both the octree structure and the corresponding geometry features. Benefiting from the inherent structure of the octree, we introduce an octree-structure-aware attention mechanism that selectively attends to the most relevant features. Furthermore, a CFG-based sampling strategy is proposed to enhance the quality and diversity of generation. Extensive experiments demonstrate that Grow3D outperforms state-of-the-art methods in both 3D shape generation quality and speed, enabling some real-time downstream applications, such as interactive 3D editing.
- 9 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (6) , 7524-7531
FlowSSC: Universal Generative Monocular Semantic Scene Completion via One-Step Latent Diffusion
Semantic Scene Completion (SSC) from monocular RGB images is a fundamental yet challenging task due to the inherent ambiguity of inferring occluded 3D geometry from a single view. While feed-forward methods have made progress, they often struggle to generate plausible details in occluded regions and preserve the fundamental spatial relationships of objects. Such accurate generative reasoning capability for the entire 3D space is critical in real-world applications. In this paper, we present FlowSSC, the first generative framework applied directly to monocular semantic scene completion. FlowSSC treats the SSC task as a conditional generation problem and can seamlessly integrate with existing feed-forward SSC methods to significantly boost their performance. To achieve real-time inference without compromising quality, we introduce Shortcut Flow-matching that operates in a compact triplane latent space. Unlike standard diffusion models that require hundreds of steps, our method utilizes a shortcut mechanism to achieve high-fidelity generation in a single step, enabling practical deployment in autonomous systems. Extensive experiments on SemanticKITTI demonstrate that FlowSSC achieves state-of-the-art performance, significantly outperforming existing baselines.
- 9 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (6) , 7524-7531
FlowSSC: Universal Generative Monocular Semantic Scene Completion via One-Step Latent Diffusion
Semantic Scene Completion (SSC) from monocular RGB images is a fundamental yet challenging task due to the inherent ambiguity of inferring occluded 3D geometry from a single view. While feed-forward methods have made progress, they often struggle to generate plausible details in occluded regions and preserve the fundamental spatial relationships of objects. Such accurate generative reasoning capability for the entire 3D space is critical in real-world applications. In this paper, we present FlowSSC, the first generative framework applied directly to monocular semantic scene completion. FlowSSC treats the SSC task as a conditional generation problem and can seamlessly integrate with existing feed-forward SSC methods to significantly boost their performance. To achieve real-time inference without compromising quality, we introduce Shortcut Flow-matching that operates in a compact triplane latent space. Unlike standard diffusion models that require hundreds of steps, our method utilizes a shortcut mechanism to achieve high-fidelity generation in a single step, enabling practical deployment in autonomous systems. Extensive experiments on SemanticKITTI demonstrate that FlowSSC achieves state-of-the-art performance, significantly outperforming existing baselines.
- Open Access
- 27 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-12
RELATE3D: REfocusing Latent Adapter for Targeted local Enhancement and Editing in 3D Generation
- 27 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-12
Open AccessRELATE3D: REfocusing Latent Adapter for Targeted local Enhancement and Editing in 3D Generation
- Open Access
- 3 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-12
DIScene: Object Decoupling and Interaction Modeling for Complex Scene Generation
- 3 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-12
Open AccessDIScene: Object Decoupling and Interaction Modeling for Complex Scene Generation
- Open Access
- 3 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-12
FragmentDiff: A Diffusion Model for Fractured Object Assembly
- 3 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-12
Open AccessFragmentDiff: A Diffusion Model for Fractured Object Assembly
- Open Access
- 3 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-11
EVSplitting: An Efficient and Visually Consistent Splitting Algorithm for 3D Gaussian Splatting
- 3 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-11
Open AccessEVSplitting: An Efficient and Visually Consistent Splitting Algorithm for 3D Gaussian Splatting
- 6 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Visualization and Computer Graphics
- Vol. 31 (9) , 5188-5198
SN2eRF: A Framework for Neural Radiance Fields Given Sparse and Noisy Poses
Neural Radiance Fields (NeRFs) have shown impressive capabilities in synthesizing photorealistic novel views. However, their application to room-size scenes is limited by the requirement of several hundred views with accurate poses for training. To address this challenge, we propose SN$^{2}$ eRF, a framework which can reconstruct the neural radiance field with significantly fewer views and noisy poses by exploiting multiple priors. Our key insight is to leverage both multi-view and monocular priors to constrain the optimization of NeRF in the setting of sparse and noisy pose inputs. Specifically, we extract and match key points to constrain pose optimization and use Ray Transformer with a monocular depth estimator to provide dense depth prior for geometry optimization. Benefiting from these priors, our approach achieves state-of-the-art accuracy in novel view synthesis for indoor room scenarios.- 6 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Visualization and Computer Graphics
- Vol. 31 (9) , 5188-5198
SN2eRF: A Framework for Neural Radiance Fields Given Sparse and Noisy Poses
Neural Radiance Fields (NeRFs) have shown impressive capabilities in synthesizing photorealistic novel views. However, their application to room-size scenes is limited by the requirement of several hundred views with accurate poses for training. To address this challenge, we propose SN$^{2}$ eRF, a framework which can reconstruct the neural radiance field with significantly fewer views and noisy poses by exploiting multiple priors. Our key insight is to leverage both multi-view and monocular priors to constrain the optimization of NeRF in the setting of sparse and noisy pose inputs. Specifically, we extract and match key points to constrain pose optimization and use Ray Transformer with a monocular depth estimator to provide dense depth prior for geometry optimization. Benefiting from these priors, our approach achieves state-of-the-art accuracy in novel view synthesis for indoor room scenarios.- Open Access
- 1 December 2023
- Published by Tsinghua University Press in Computational Visual Media
- Vol. 9 (4) , 687-697
Neural 3D reconstruction from sparse views using geometric priors
Sparse view 3D reconstruction has attracted increasing attention with the development of neural implicit 3D representation. Existing methods usually only make use of 2D views, requiring a dense set of input views for accurate 3D reconstruction. In this paper, we show that accurate 3D reconstruction can be achieved by incorporating geometric priors into neural implicit 3D reconstruction. Our method adopts the signed distance function as the 3D representation, and learns a generalizable 3D surface reconstruction model from sparse views. Specifically, we build a more effective and sparse feature volume from the input views by using corresponding depth maps, which can be provided by depth sensors or directly predicted from the input views. We recover better geometric details by imposing both depth and surface normal constraints in addition to the color loss when training the neural implicit 3D representation. Experiments demonstrate that our method both outperforms state-of-the-art approaches, and achieves good generalizability.
- 1 December 2023
- Published by Tsinghua University Press in Computational Visual Media
- Vol. 9 (4) , 687-697
Open AccessNeural 3D reconstruction from sparse views using geometric priors
Sparse view 3D reconstruction has attracted increasing attention with the development of neural implicit 3D representation. Existing methods usually only make use of 2D views, requiring a dense set of input views for accurate 3D reconstruction. In this paper, we show that accurate 3D reconstruction can be achieved by incorporating geometric priors into neural implicit 3D reconstruction. Our method adopts the signed distance function as the 3D representation, and learns a generalizable 3D surface reconstruction model from sparse views. Specifically, we build a more effective and sparse feature volume from the input views by using corresponding depth maps, which can be provided by depth sensors or directly predicted from the input views. We recover better geometric details by imposing both depth and surface normal constraints in addition to the color loss when training the neural implicit 3D representation. Experiments demonstrate that our method both outperforms state-of-the-art approaches, and achieves good generalizability.
- 17 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 32 (10577149) , 6401-6412
SPS: Accurate and Real-Time Semantic Positioning System Based on Low-Cost DEM Maps
This paper presents a Semantic Positioning System (SPS) to enhance the accuracy of mobile device geo-localization in outdoor urban environments. Although the traditional Global Positioning System (GPS) can offer a rough localization, it lacks the necessary accuracy for applications such as Augmented Reality (AR). Our SPS integrates Geographic Information System (GIS) data, GPS signals, and visual image information to estimate the 6 Degree-of-Freedom (DoF) pose through cross-view semantic matching. This approach has excellent scalability to support GIS context with Levels of Detail (LOD). The map data representation is Digital Elevation Model (DEM), a cost-effective aerial map that allows for fast deployment for large-scale areas. However, the DEM lacks geometric and texture details, making it challenging for traditional visual feature extraction to establish pixel/voxel level cross-view correspondences. To address this, we sample observation pixels from the query ground-view image using predicted semantic labels. We then propose an iterative homography estimation method with semantic correspondences. To improve the efficiency of the overall system, we further employ a heuristic search to speedup the matching process. The proposed method is robust, real-time, and automatic. Quantitative experiments on the challenging Bund dataset show that we achieve a positioning accuracy of 73.24%, surpassing the baseline skyline-based method by 20%. Compared with the state-of-the-art semantic-based approach on the Kitti dataset, we improve the positioning accuracy by an average 5%.
- 17 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 32 (10577149) , 6401-6412
SPS: Accurate and Real-Time Semantic Positioning System Based on Low-Cost DEM Maps
This paper presents a Semantic Positioning System (SPS) to enhance the accuracy of mobile device geo-localization in outdoor urban environments. Although the traditional Global Positioning System (GPS) can offer a rough localization, it lacks the necessary accuracy for applications such as Augmented Reality (AR). Our SPS integrates Geographic Information System (GIS) data, GPS signals, and visual image information to estimate the 6 Degree-of-Freedom (DoF) pose through cross-view semantic matching. This approach has excellent scalability to support GIS context with Levels of Detail (LOD). The map data representation is Digital Elevation Model (DEM), a cost-effective aerial map that allows for fast deployment for large-scale areas. However, the DEM lacks geometric and texture details, making it challenging for traditional visual feature extraction to establish pixel/voxel level cross-view correspondences. To address this, we sample observation pixels from the query ground-view image using predicted semantic labels. We then propose an iterative homography estimation method with semantic correspondences. To improve the efficiency of the overall system, we further employ a heuristic search to speedup the matching process. The proposed method is robust, real-time, and automatic. Quantitative experiments on the challenging Bund dataset show that we achieve a positioning accuracy of 73.24%, surpassing the baseline skyline-based method by 20%. Compared with the state-of-the-art semantic-based approach on the Kitti dataset, we improve the positioning accuracy by an average 5%.
- 11 November 2022
- Published by Springer Nature
- p. 506-522
CIRCLE: Convolutional Implicit Reconstruction and Completion for Large-Scale Indoor Scene
- 11 November 2022
- Published by Springer Nature
- p. 506-522
CIRCLE: Convolutional Implicit Reconstruction and Completion for Large-Scale Indoor Scene