计算机视觉，AI+

招收人工智能（智能科学与技术）、计算机科学与技术、软件工程等相关专业的考生

（1）教育部重大人才计划特聘教授

（2）国家网信专项人才

（3）中国科学院青年创新促进会优秀会员

（4）北京市科技新星

代表性论文：

(1) PanoGS: Panorama-based 3D Scene Stylization with Style and Geometry Consistency. IEEE Transactions on Image Processing, 2026

(2) GW-VRWKV: Synergizing Global and Window-Based Local Perception in VRWKV for Hyperspectral Image Classification. IEEE Transactions on Circuits and Systems for Video Technology,2026

(3) VGR: Visual Grounded Reasoning [C]. ICLR 2026

(4) Part-Pose: Balancing Specificity with Consistency in Part-Level Feature for Accurate and Robust Category-Level Object Pose Estimation. IEEE Transactions on Circuits and Systems for Video Technology,2026

(5) Pose-Free Omnidirectional Gaussian Splatting for 360-Degree Videos with Consistent Depth Priors [C]. CVPR 2026

(6) Hi-RWKV: Hierarchical RWKV Modeling for Hyperspectral Image Classification [J]. IEEE Transactions on Image Processing, 2026

(7) Cross-Domain Incremental Image Classification via Domain-specific Prototypes and Contrastive Learning [J]. IEEE Transactions on Geoscience and Remote Sensing,2026

(8) PS-CAD: Local Geometry Guidance via Prompting and Selection for CAD Reconstruction [J]. ACM Transactions on Graphics, 2025

(9) Empowering Vector Graphics with Consistently Arbitrary Viewing and View-dependent Visibility[C]. CVPR 2025

(10) D3CTTA: Domain-Dependent Decorrelation for Continual Test-Time Adaption of 3D LiDAR Segmentation[C]. CVPR 2025

(11) Activating Sparse Part Concepts for 3D Class Incremental Learning[C]. CVPR 2025

(12) BGPSeg: Boundary-Guided Primitive Instance Segmentation of Point Clouds [J]. IEEE Transactions on Image Processing, 2025

(13) MambaHSI+: Multidirectional State Propagation for Efficient Hyperspectral Image Classification [J]. IEEE Transactions on Geoscience and Remote Sensing,2025

(14) Efficient Single Correspondence Voting for Point Cloud Registration [J]. IEEE Transactions on Image Processing, 2024

(15) Multi-view Human Pose and Shape Estimation via Mesh-aligned Voxel Interpolation [J], Information Fusion, 2024

(16) FC-4DFS: Frequency-controlled Flexible 4D Facial Expression Synthesizing [C]. ACM MM 2024

(17) World to Code: Multi-modal Data Generation via Self-Instructed Compositional Captioning and Filtering. EMNLP 2024

(18) SPDET: Edge-Aware Self-Supervised Panoramic Depth Estimation Transformer With Spherical Geometry [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023

(19) PartCom: Part Composition Learning for 3D Open-Set Recognition [J]. International Journal of Computer Vision, 2023

(20) NR-MVSNet: Learning Multi-view Stereo based on Normal Consistency and Depth Refinement [J]. IEEE Transactions on Image Processing, 2023

(21) Decompose Novel into Known: Part Concept Learning For 3D Novel Class Discovery [C]. NeurIPS, 2023

(22) VectorFloorSeg: Two-Stream Graph Attention Network for Vectorized Roughcast Floorplan Segmentation[C], CVPR 2023 (HIGHLIGHT paper)

(23) SSF: Accelerating Training of Spiking Neural Networks with Stabilized Spiking Flow [C]. ICCV 2023

(24) Progressive Graph Matching Network for Correspondences[C]. Pacific Graphics, 2023

(25) Landmark Guided 4D Facial Expression Generation[C]. SIGGRAPH Asia, 2023

(26) DS-MVSNet: Unsupervised Multi-view Stereo via Depth Synthesis[C], ACM Multimedia, 2022

(27) ACDNet: Adaptively Combined Dilated Convolution for Monocular Panorama Depth Estimation[C], AAAI, 2022

(28) Context-Aware 3D Point Cloud Semantic Segmentation With Plane Guidance[J], IEEE Transactions on Multimedia, 2022

(29) MEMformer: Transformer-based 3D Human Motion Estimation from MoCap Markers[C], SIGGRAPH Asia, 2022

(30) Registration Method for Point Clouds of Complex Rock-mass Based on Dual Structure Information[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022

(31) Joint Specular Highlight Detection and Removal in Single Images via Unet-Transformer[J], Computational Visual Media, 2022

(32) Single-Image Specular Highlight Removal via Real-World Dataset Construction[J]. IEEE Transactions on Multimedia, 2021

(33) A Novel Rock-Mass Point Cloud Registration Method Based on Feature Line Extraction and Feature Point Matching [J]. IEEE Transactions on Geoscience and Remote Sensing, 2021

(34) Efficient Pairwise 3D Registration of Urban Scenes Via Hybrid Structural Descriptors [J]. IEEE Transactions on Geoscience and Remote Sensing, 2021

(35) Data-driven Floor Plan Understanding in Rural Residential Buildings via Deep Recognition [J], Information Sciences, 2021

(36) End-to-end 3D Point Cloud Instance Segmentation without Detection[C]. CVPR 2020

(37) Blending Surface Segmentation and Editing for 3D Models [J], IEEE Transactions on Visualization and Computer Graphics, 2020

(38) Extracting Cycle-Aware Feature Curve Networks from 3D Models [J], Computer-Aided Design, 2020

(39) Accurate Rock-mass Extraction from Terrestrial Laser Point Clouds via Multi-scale and Multi-view Convolutional Feature Representation [J]. IEEE Transactions on Geoscience and Remote Sensing, 2020

(40) Efficient Rock-Mass Point Cloud Registration Using n-Point Complete Graphs [J]. IEEE Transactions on Geoscience and Remote Sensing, 2020

(41) CIAN: Cross-Image Affinity Net for Weakly Supervised Semantic Segmentation [C]，AAAI, 2020