Hao Jiang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Hao Jiang has a total of 73 co-authors. Here are the top co-authors by number of shared publications:
- Zhaoqi Wang
Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China
8 shared publications - Zhaoxin Li
Agricultural Information Institute, Chinese Academy of Agricultural Sciences, Beijing 100081, China
4 shared publications Jianing Chen
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
3 shared publicationsYujun Cai
The University of Queensland
3 shared publications- Zhigang Deng
Department of Computer Science, University of Houston, Houston, TX, USA
3 shared publications Honglong Zhao
Institute of Computing Technology Chinese Academy of Sciences ICT
2 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 5690-5699
STDR: Spatio-Temporal Decoupling for Real-Time Dynamic Scene Rendering
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 5690-5699
Open AccessSTDR: Spatio-Temporal Decoupling for Real-Time Dynamic Scene Rendering
- Open Access
Polymap: generating high definition map based on rasterized polygons
The perception of high-definition maps is an integral component of environmental perception in autonomous driving systems. Existing research have often focused on online construction of high-definition maps. For instance, the Maptr[9] series employ a detection-based method to output vectorized map instances parallelly in an end-to-end manner. However, despite their capability for real-time construction, detection-based methods are observed to lack robust generalizability[19], which hampers their applicability in auto-labeling systems. Therefore, aiming to improve the generalizability, we reinterpret road elements as rasterized polygons and design a concise framework based on instance segmentation. Initially, a segmentation-based transformer is employed to deliver instance masks in an end-to-end manner; succeeding this step, a Potrace-based[17] post-processing module is used to ultimately yield vectorized map elements. Quantitative results attained on the Nuscene[1] dataset substantiate the effectiveness and generaliz-ability of our method.
Open AccessPolymap: generating high definition map based on rasterized polygons
The perception of high-definition maps is an integral component of environmental perception in autonomous driving systems. Existing research have often focused on online construction of high-definition maps. For instance, the Maptr[9] series employ a detection-based method to output vectorized map instances parallelly in an end-to-end manner. However, despite their capability for real-time construction, detection-based methods are observed to lack robust generalizability[19], which hampers their applicability in auto-labeling systems. Therefore, aiming to improve the generalizability, we reinterpret road elements as rasterized polygons and design a concise framework based on instance segmentation. Initially, a segmentation-based transformer is employed to deliver instance masks in an end-to-end manner; succeeding this step, a Potrace-based[17] post-processing module is used to ultimately yield vectorized map elements. Quantitative results attained on the Nuscene[1] dataset substantiate the effectiveness and generaliz-ability of our method.
- 4 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36  (4) , 4212-4225
DVP-MVS++: Synergize Depth-Normal-Edge and Harmonized Visibility Prior for Multi-View Stereo
Recently, patch deformation-based methods have demonstrated significant effectiveness in multi-view stereo due to their incorporation of deformable and expandable perception for reconstructing textureless areas. However, these methods generally focus on identifying reliable pixel correlations to mitigate matching ambiguity of patch deformation, while neglecting the deformation instability caused by edge-skipping and visibility occlusions, which may cause potential estimation deviations. To address these issues, we propose DVP-MVS++, an innovative approach that synergizes both depth-normal-edge aligned and harmonized cross-view priors for robust and visibility-aware patch deformation. Specifically, to avoid edge-skipping, we first apply DepthPro, Metric3Dv2 and Roberts operator to generate coarse depth maps, normal maps and edge maps, respectively. These maps are then aligned via an erosion-dilation strategy to produce fine-grained homogeneous boundaries for facilitating robust patch deformation. Moreover, we reformulate view selection weights as visibility maps, and then implement both an enhanced cross-view depth reprojection and an area-maximization strategy to help reliably restore visible areas and effectively balance deformed patch. Additionally, we obtain geometry consistency by adopting both aggregated normals via view selection and projection depth differences via epipolar lines, and then employ SHIQ for highlight correction to facilitate highlight perception capacity, thus improving reconstruction quality during propagation and refinement stage. Evaluations on ETH3D, Tanks & Temples and Strecha datasets exhibit the state-of-the-art performance and robust generalization capability of our proposed method.
- 4 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36  (4) , 4212-4225
DVP-MVS++: Synergize Depth-Normal-Edge and Harmonized Visibility Prior for Multi-View Stereo
Recently, patch deformation-based methods have demonstrated significant effectiveness in multi-view stereo due to their incorporation of deformable and expandable perception for reconstructing textureless areas. However, these methods generally focus on identifying reliable pixel correlations to mitigate matching ambiguity of patch deformation, while neglecting the deformation instability caused by edge-skipping and visibility occlusions, which may cause potential estimation deviations. To address these issues, we propose DVP-MVS++, an innovative approach that synergizes both depth-normal-edge aligned and harmonized cross-view priors for robust and visibility-aware patch deformation. Specifically, to avoid edge-skipping, we first apply DepthPro, Metric3Dv2 and Roberts operator to generate coarse depth maps, normal maps and edge maps, respectively. These maps are then aligned via an erosion-dilation strategy to produce fine-grained homogeneous boundaries for facilitating robust patch deformation. Moreover, we reformulate view selection weights as visibility maps, and then implement both an enhanced cross-view depth reprojection and an area-maximization strategy to help reliably restore visible areas and effectively balance deformed patch. Additionally, we obtain geometry consistency by adopting both aggregated normals via view selection and projection depth differences via epipolar lines, and then employ SHIQ for highlight correction to facilitate highlight perception capacity, thus improving reconstruction quality during propagation and refinement stage. Evaluations on ETH3D, Tanks & Temples and Strecha datasets exhibit the state-of-the-art performance and robust generalization capability of our proposed method.
- Open Access
From Tokens to Nodes: Semantic-Guided Motion Control for Dynamic 3D Gaussian Splatting
Dynamic 3D reconstruction from monocular videos remains difficult due to the ambiguity inferring 3D motion from limited views and computational demands of modeling temporally varying scenes. While recent sparse control methods alleviate computation by reducing millions of Gaussians to thousands of control points, they suffer from a critical limitation: they allocate points purely by geometry, leading to static redundancy and dynamic insufficiency. We propose a motion-adaptive framework that aligns control density with motion complexity. Leveraging semantic and motion priors from vision foundation models, we establish patch-token-node correspondences and apply motion-adaptive compression to concentrate control points in dynamic regions while suppressing redundancy in static backgrounds. Our approach achieves flexible representational density adaptation through iterative voxelization and motion tendency scoring, directly addressing the fundamental mismatch between control point allocation and motion complexity. To capture temporal evolution, we introduce spline-based trajectory parameterization initialized by 2D tracklets, replacing MLP-based deformation fields to achieve smoother motion representation and more stable optimization. Extensive experiments demonstrate significant improvements in reconstruction quality and efficiency over existing state-of-the-art methods.
Open AccessFrom Tokens to Nodes: Semantic-Guided Motion Control for Dynamic 3D Gaussian Splatting
Dynamic 3D reconstruction from monocular videos remains difficult due to the ambiguity inferring 3D motion from limited views and computational demands of modeling temporally varying scenes. While recent sparse control methods alleviate computation by reducing millions of Gaussians to thousands of control points, they suffer from a critical limitation: they allocate points purely by geometry, leading to static redundancy and dynamic insufficiency. We propose a motion-adaptive framework that aligns control density with motion complexity. Leveraging semantic and motion priors from vision foundation models, we establish patch-token-node correspondences and apply motion-adaptive compression to concentrate control points in dynamic regions while suppressing redundancy in static backgrounds. Our approach achieves flexible representational density adaptation through iterative voxelization and motion tendency scoring, directly addressing the fundamental mismatch between control point allocation and motion complexity. To capture temporal evolution, we introduce spline-based trajectory parameterization initialized by 2D tracklets, replacing MLP-based deformation fields to achieve smoother motion representation and more stable optimization. Extensive experiments demonstrate significant improvements in reconstruction quality and efficiency over existing state-of-the-art methods.
- 28 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 35  (11) , 11244-11257
SED-MVS: Segmentation-Driven and Edge-Aligned Deformation Multi-View Stereo With Depth Restoration and Occlusion Constraint
Recently, patch-deformation methods have exhibited significant effectiveness in multi-view stereo owing to the deformable and expandable patches in reconstructing textureless areas. However, existing approaches neglect to address the problem of deformation instability caused by easily overlooked edge-skipping, potentially leading to matching distortions, thus leaving room for further improvement. To fill this gap, we propose SED-MVS, which adopts panoptic segmentation and multi-trajectory diffusion strategy for segmentation-driven and edge-aligned patch deformation. Specifically, to prevent unanticipated edge-skipping, we first employ SAM2 for panoptic segmentation as depth-edge guidance to guide patch deformation, followed by multi-trajectory diffusion strategy to ensure patches are comprehensively aligned with depth edges. Moreover, to avoid potential inaccuracy of random initialization, we combine both sparse points from LoFTR and monocular depth map from DepthAnything V2 to restore reliable and realistic depth map for initialization and supervised guidance. Finally, we integrate the segmentation image with the monocular depth map to exploit inter-instance occlusion relationship, then further regard them as occlusion map to implement two distinct edge constraint, thereby facilitating occlusion-aware patch deformation. Extensive results on ETH3D, Tanks & Temples, BlendedMVS, Strecha and DL3DV-10K datasets validate the state-of-the-art performance and robust generalization capability of our proposed method.
- 28 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 35  (11) , 11244-11257
SED-MVS: Segmentation-Driven and Edge-Aligned Deformation Multi-View Stereo With Depth Restoration and Occlusion Constraint
Recently, patch-deformation methods have exhibited significant effectiveness in multi-view stereo owing to the deformable and expandable patches in reconstructing textureless areas. However, existing approaches neglect to address the problem of deformation instability caused by easily overlooked edge-skipping, potentially leading to matching distortions, thus leaving room for further improvement. To fill this gap, we propose SED-MVS, which adopts panoptic segmentation and multi-trajectory diffusion strategy for segmentation-driven and edge-aligned patch deformation. Specifically, to prevent unanticipated edge-skipping, we first employ SAM2 for panoptic segmentation as depth-edge guidance to guide patch deformation, followed by multi-trajectory diffusion strategy to ensure patches are comprehensively aligned with depth edges. Moreover, to avoid potential inaccuracy of random initialization, we combine both sparse points from LoFTR and monocular depth map from DepthAnything V2 to restore reliable and realistic depth map for initialization and supervised guidance. Finally, we integrate the segmentation image with the monocular depth map to exploit inter-instance occlusion relationship, then further regard them as occlusion map to implement two distinct edge constraint, thereby facilitating occlusion-aware patch deformation. Extensive results on ETH3D, Tanks & Temples, BlendedMVS, Strecha and DL3DV-10K datasets validate the state-of-the-art performance and robust generalization capability of our proposed method.
- Open Access
STDR: Spatio-Temporal Decoupling for Real-Time Dynamic Scene Rendering
Although dynamic scene reconstruction has long been a fundamental challenge in 3D vision, the recent emergence of 3D Gaussian Splatting (3DGS) offers a promising direction by enabling high-quality, real-time rendering through explicit Gaussian primitives. However, existing 3DGS-based methods for dynamic reconstruction often suffer from \textit{spatio-temporal incoherence} during initialization, where canonical Gaussians are constructed by aggregating observations from multiple frames without temporal distinction. This results in spatio-temporally entangled representations, making it difficult to model dynamic motion accurately. To overcome this limitation, we propose \textbf{STDR} (Spatio-Temporal Decoupling for Real-time rendering), a plug-and-play module that learns spatio-temporal probability distributions for each Gaussian. STDR introduces a spatio-temporal mask, a separated deformation field, and a consistency regularization to jointly disentangle spatial and temporal patterns. Extensive experiments demonstrate that incorporating our module into existing 3DGS-based dynamic scene reconstruction frameworks leads to notable improvements in both reconstruction quality and spatio-temporal consistency across synthetic and real-world benchmarks.
Open AccessSTDR: Spatio-Temporal Decoupling for Real-Time Dynamic Scene Rendering
Although dynamic scene reconstruction has long been a fundamental challenge in 3D vision, the recent emergence of 3D Gaussian Splatting (3DGS) offers a promising direction by enabling high-quality, real-time rendering through explicit Gaussian primitives. However, existing 3DGS-based methods for dynamic reconstruction often suffer from \textit{spatio-temporal incoherence} during initialization, where canonical Gaussians are constructed by aggregating observations from multiple frames without temporal distinction. This results in spatio-temporally entangled representations, making it difficult to model dynamic motion accurately. To overcome this limitation, we propose \textbf{STDR} (Spatio-Temporal Decoupling for Real-time rendering), a plug-and-play module that learns spatio-temporal probability distributions for each Gaussian. STDR introduces a spatio-temporal mask, a separated deformation field, and a consistency regularization to jointly disentangle spatial and temporal patterns. Extensive experiments demonstrate that incorporating our module into existing 3DGS-based dynamic scene reconstruction frameworks leads to notable improvements in both reconstruction quality and spatio-temporal consistency across synthetic and real-world benchmarks.
- Open Access
SED-MVS: Segmentation-Driven and Edge-Aligned Deformation Multi-View Stereo with Depth Restoration and Occlusion Constraint
Recently, patch-deformation methods have exhibited significant effectiveness in multi-view stereo owing to the deformable and expandable patches in reconstructing textureless areas. However, such methods primarily emphasize broadening the receptive field in textureless areas, while neglecting deformation instability caused by easily overlooked edge-skipping, potentially leading to matching distortions. To address this, we propose SED-MVS, which adopts panoptic segmentation and multi-trajectory diffusion strategy for segmentation-driven and edge-aligned patch deformation. Specifically, to prevent unanticipated edge-skipping, we first employ SAM2 for panoptic segmentation as depth-edge guidance to guide patch deformation, followed by multi-trajectory diffusion strategy to ensure patches are comprehensively aligned with depth edges. Moreover, to avoid potential inaccuracy of random initialization, we combine both sparse points from LoFTR and monocular depth map from DepthAnything V2 to restore reliable and realistic depth map for initialization and supervised guidance. Finally, we integrate segmentation image with monocular depth map to exploit inter-instance occlusion relationship, then further regard them as occlusion map to implement two distinct edge constraint, thereby facilitating occlusion-aware patch deformation. Extensive results on ETH3D, Tanks & Temples, BlendedMVS and Strecha datasets validate the state-of-the-art performance and robust generalization capability of our proposed method.
Open AccessSED-MVS: Segmentation-Driven and Edge-Aligned Deformation Multi-View Stereo with Depth Restoration and Occlusion Constraint
Recently, patch-deformation methods have exhibited significant effectiveness in multi-view stereo owing to the deformable and expandable patches in reconstructing textureless areas. However, such methods primarily emphasize broadening the receptive field in textureless areas, while neglecting deformation instability caused by easily overlooked edge-skipping, potentially leading to matching distortions. To address this, we propose SED-MVS, which adopts panoptic segmentation and multi-trajectory diffusion strategy for segmentation-driven and edge-aligned patch deformation. Specifically, to prevent unanticipated edge-skipping, we first employ SAM2 for panoptic segmentation as depth-edge guidance to guide patch deformation, followed by multi-trajectory diffusion strategy to ensure patches are comprehensively aligned with depth edges. Moreover, to avoid potential inaccuracy of random initialization, we combine both sparse points from LoFTR and monocular depth map from DepthAnything V2 to restore reliable and realistic depth map for initialization and supervised guidance. Finally, we integrate segmentation image with monocular depth map to exploit inter-instance occlusion relationship, then further regard them as occlusion map to implement two distinct edge constraint, thereby facilitating occlusion-aware patch deformation. Extensive results on ETH3D, Tanks & Temples, BlendedMVS and Strecha datasets validate the state-of-the-art performance and robust generalization capability of our proposed method.
- Open Access
COMET: Benchmark for Comprehensive Biological Multi-omics Evaluation Tasks and Language Models
As key elements within the central dogma, DNA, RNA, and proteins play crucial roles in maintaining life by guaranteeing accurate genetic expression and implementation. Although research on these molecules has profoundly impacted fields like medicine, agriculture, and industry, the diversity of machine learning approaches-from traditional statistical methods to deep learning models and large language models-poses challenges for researchers in choosing the most suitable models for specific tasks, especially for cross-omics and multi-omics tasks due to the lack of comprehensive benchmarks. To address this, we introduce the first comprehensive multi-omics benchmark COMET (Benchmark for Biological COmprehensive Multi-omics Evaluation Tasks and Language Models), designed to evaluate models across single-omics, cross-omics, and multi-omics tasks. First, we curate and develop a diverse collection of downstream tasks and datasets covering key structural and functional aspects in DNA, RNA, and proteins, including tasks that span multiple omics levels. Then, we evaluate existing foundational language models for DNA, RNA, and proteins, as well as the newly proposed multi-omics method, offering valuable insights into their performance in integrating and analyzing data from different biological modalities. This benchmark aims to define critical issues in multi-omics research and guide future directions, ultimately promoting advancements in understanding biological processes through integrated and different omics data analysis.
Open AccessCOMET: Benchmark for Comprehensive Biological Multi-omics Evaluation Tasks and Language Models
As key elements within the central dogma, DNA, RNA, and proteins play crucial roles in maintaining life by guaranteeing accurate genetic expression and implementation. Although research on these molecules has profoundly impacted fields like medicine, agriculture, and industry, the diversity of machine learning approaches-from traditional statistical methods to deep learning models and large language models-poses challenges for researchers in choosing the most suitable models for specific tasks, especially for cross-omics and multi-omics tasks due to the lack of comprehensive benchmarks. To address this, we introduce the first comprehensive multi-omics benchmark COMET (Benchmark for Biological COmprehensive Multi-omics Evaluation Tasks and Language Models), designed to evaluate models across single-omics, cross-omics, and multi-omics tasks. First, we curate and develop a diverse collection of downstream tasks and datasets covering key structural and functional aspects in DNA, RNA, and proteins, including tasks that span multiple omics levels. Then, we evaluate existing foundational language models for DNA, RNA, and proteins, as well as the newly proposed multi-omics method, offering valuable insights into their performance in integrating and analyzing data from different biological modalities. This benchmark aims to define critical issues in multi-omics research and guide future directions, ultimately promoting advancements in understanding biological processes through integrated and different omics data analysis.
- Open Access
GradiSeg: Gradient-Guided Gaussian Segmentation with Enhanced 3D Boundary Precision
While 3D Gaussian Splatting enables high-quality real-time rendering, existing Gaussian-based frameworks for 3D semantic segmentation still face significant challenges in boundary recognition accuracy. To address this, we propose a novel 3DGS-based framework named GradiSeg, incorporating Identity Encoding to construct a deeper semantic understanding of scenes. Our approach introduces two key modules: Identity Gradient Guided Densification (IGD) and Local Adaptive K-Nearest Neighbors (LA-KNN). The IGD module supervises gradients of Identity Encoding to refine Gaussian distributions along object boundaries, aligning them closely with boundary contours. Meanwhile, the LA-KNN module employs position gradients to adaptively establish locality-aware propagation of Identity Encodings, preventing irregular Gaussian spreads near boundaries. We validate the effectiveness of our method through comprehensive experiments. Results show that GradiSeg effectively addresses boundary-related issues, significantly improving segmentation accuracy without compromising scene reconstruction quality. Furthermore, our method's robust segmentation capability and decoupled Identity Encoding representation make it highly suitable for various downstream scene editing tasks, including 3D object removal, swapping and so on.
Open AccessGradiSeg: Gradient-Guided Gaussian Segmentation with Enhanced 3D Boundary Precision
While 3D Gaussian Splatting enables high-quality real-time rendering, existing Gaussian-based frameworks for 3D semantic segmentation still face significant challenges in boundary recognition accuracy. To address this, we propose a novel 3DGS-based framework named GradiSeg, incorporating Identity Encoding to construct a deeper semantic understanding of scenes. Our approach introduces two key modules: Identity Gradient Guided Densification (IGD) and Local Adaptive K-Nearest Neighbors (LA-KNN). The IGD module supervises gradients of Identity Encoding to refine Gaussian distributions along object boundaries, aligning them closely with boundary contours. Meanwhile, the LA-KNN module employs position gradients to adaptively establish locality-aware propagation of Identity Encodings, preventing irregular Gaussian spreads near boundaries. We validate the effectiveness of our method through comprehensive experiments. Results show that GradiSeg effectively addresses boundary-related issues, significantly improving segmentation accuracy without compromising scene reconstruction quality. Furthermore, our method's robust segmentation capability and decoupled Identity Encoding representation make it highly suitable for various downstream scene editing tasks, including 3D object removal, swapping and so on.
- Open Access
- 19 November 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 43  (6) , 1-12
Large Scale Farm Scene Modeling from Remote Sensing Imagery
In this paper we propose a scalable framework for large-scale farm scene modeling that utilizes remote sensing data, specifically satellite images. Our approach begins by accurately extracting and categorizing the distributions of various scene elements from satellite images into four distinct layers: fields, trees, roads, and grasslands. For each layer, we introduce a set of controllable Parametric Layout Models (PLMs). These models are capable of learning layout parameters from satellite images, enabling them to generate complex, large-scale farm scenes that closely reproduce reality across multiple scales. Additionally, our framework provides intuitive control for users to adjust layout parameters to simulate different stages of crop growth and planting patterns. This adaptability makes our model an excellent tool for graphics and virtual reality applications. Experimental results demonstrate that our approach can rapidly generate a variety of realistic and highly detailed farm scenes with minimal inputs.
- 19 November 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 43  (6) , 1-12
Open AccessLarge Scale Farm Scene Modeling from Remote Sensing Imagery
In this paper we propose a scalable framework for large-scale farm scene modeling that utilizes remote sensing data, specifically satellite images. Our approach begins by accurately extracting and categorizing the distributions of various scene elements from satellite images into four distinct layers: fields, trees, roads, and grasslands. For each layer, we introduce a set of controllable Parametric Layout Models (PLMs). These models are capable of learning layout parameters from satellite images, enabling them to generate complex, large-scale farm scenes that closely reproduce reality across multiple scales. Additionally, our framework provides intuitive control for users to adjust layout parameters to simulate different stages of crop growth and planting patterns. This adaptability makes our model an excellent tool for graphics and virtual reality applications. Experimental results demonstrate that our approach can rapidly generate a variety of realistic and highly detailed farm scenes with minimal inputs.