Research Scientist
InnoPeak Technology, Inc (OPPO US Research Center).

Email: yanqinganssg at gmail dot com
Office: 2479 E Bayshore Rd #110, Palo Alto, CA 94303
Links: GitHub | LinkedIn | Google Scholar

I am a staff research scientist & tech lead at the AI-wearable devices department of InnoPeak Technology (a.k.a., OPPO US Research Center), leading a team focusing on 3D environment perception and LLM-powered user-device interaction algorithms design. From 2017 to 2021, I served as a senior research scientist for 3D Vision-related algorithm development at JD.com American Technologies in Mountain View, CA. Before that, I received my Ph.D. degree in Computer Science from Wuhan University, 2017.

My interests primarily lie in the field of 3D Computer Vision and Embodied AI. In particular, I tackle the challenges in scene reconstruction, content generation, layout understanding, panoptic segmentation, SLAM, depth estimation and multimodal foundation models.

• NARUTO: Neural Active Reconstruction from Uncertain Target Observations

  Ziyue Feng^*, Huangying Zhan^*, Zheng Chen, Qingan Yan, Xiangyu Xu, Changjiang Cai, Bing Li, Qilun Zhu, Yi Xu

  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024.

  [Paper (arXiv)]  [Project Page]  [Code]

• RIAV-MVS: Recurrent-Indexing an Asymmetric Volume for Multi-View Stereo

  Changjiang Cai, Pan Ji, Qingan Yan, Yi Xu

  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023.

  [Paper]  [Project Page] [Code]

• GeoRefine: Self-Supervised Online Depth Refinement for Accurate Dense Mapping

  Pan Ji^*, Qingan Yan^*, Yuxin Ma, Yi Xu

  European Conference on Computer Vision (ECCV), 2022.

  [Paper]  [Supplemental Material]

• Point Cloud Completion via Skeleton-Detail Transformer

  Wenxiao Zhang, Huajian Zhou, Zhen Dong, Jun Liu, Qingan Yan, Chunxia Xiao

  IEEE Transactions on Visualization and Computer Graphics (TVCG), 2022.

  [Paper]  [Code]

• PlaneMVS: 3D Plane Reconstruction from Multi-View Stereo

  Jiachen Liu, Pan Ji, Nitin Bansal, Changjiang Cai, Qingan Yan, Xiaolei Huang, Yi Xu

  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022.

  [Paper]  [Code]

• Rank-PointRetrieval: Reranking Point Cloud Retrieval via a Visually Consistent Registration Evaluation

  Wenxiao Zhang, Huajian Zhou, Zhen Dong, Qingan Yan, Chunxia Xiao

  IEEE Transactions on Visualization and Computer Graphics (TVCG), 2022.

  [Paper]  [Code]

• Seamless Texture Optimization for RGB-D Reconstruction

  Yanping Fu, Qingan Yan, Jie Liao, Huajian Zhou, Jin Tang, Chunxia Xiao

  IEEE Transactions on Visualization and Computer Graphics (TVCG), 2021.

  [Paper]  [Code]

• Detail Preserved Point Cloud Completion via Separated Feature Aggregation

  Wenxiao Zhang, Qingan Yan, Chunxia Xiao

  European Conference on Computer Vision (ECCV), 2020.

  [Paper]  [Code] 

• Joint Texture and Geometry Optimization for RGB-D Reconstruction

  Yanping Fu, Qingan Yan, Jie Liao, Chunxia Xiao

  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020.

  [Paper]  [Supplemental Material]  [Implementation]

• Multi-Stage Point Completion Network with Critical Set Supervision

  Wenxiao Zhang, Chengjiang Long, Qingan Yan, Alix L.H. Chow, Chunxia Xiao

  Computer-Aided Geometric Design (CAGD), 2020. (Proceedings of Geometric Modeling and Processing 2020)

  [Paper] 

• Pyramid Multi‐View Stereo with Local Consistency

  Jie Liao, Yanping Fu, Qingan Yan, Chunxia Xiao

  Computer Graphics Forum (CGF), 2019. (Proceedings of Pacific Graphics 2019)

  [Paper]  [Code]

• Joint Bilateral Propagation Upsampling for Unstructured Multi-view Stereo

  Mengqiang Wei, Qingan Yan, Fei Luo, Chengfang Song, Chunxia Xiao

  The Visual Computer (TVC), 2019. (Proceedings of Computer Graphics International 2019)

  [Paper]  [Supplemental Material]

• Effective Shadow Removal via Multi-scale Image Decomposition

  Ling Zhang, Qingan Yan, Yao Zhu, Xiaolong Zhang, Chunxia Xiao

  The Visual Computer (TVC), 2019. (Proceedings of Computer Graphics International 2019)

  [Paper]  [Supplemental Material] [Code]

• Texture Mapping for 3D Reconstruction with RGB-D Sensor

  Yanping Fu, Qingan Yan, Long Yang, Jie Liao, Chunxia Xiao

  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018.

  [Paper]  [Implementation]

• Surface Reconstruction via Fusing Sparse-Sequence of Depth Images

  Long Yang, Qingan Yan, Yanping Fu, Chunxia Xiao

  IEEE Transactions on Visualization and Computer Graphics (TVCG), 2018.

  [Paper]

• Illumination Decomposition for Photograph with Multiple Light Sources

  Ling Zhang, Qingan Yan, Zheng Liu, Hua Zou, Chunxia Xiao

  IEEE Transactions on Image Processing (TIP), 2017.

  [Paper]  [Code]

• Distinguishing the Indistinguishable: Exploring Structural Ambiguities via Geodesic Context

  Qingan Yan, Long Yang, Ling Zhang, Chunxia Xiao

  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. (Spotlight Oral)

  [Paper]  [Code]

• Shape-Controllable Geometry Completion for Point Cloud Models

  Long Yang, Qingan Yan, Chunxia Xiao

  The Visual Computer (TVC), 2017.

  [Paper]

• Geometrically Based Linear Iterative Clustering for Quantitative Feature Correspondence

  Qingan Yan, Long Yang, Chao Liang, Huajun Liu, Ruimin Hu, Chunxia Xiao

  Computer Graphics Forum (CGF), 2016. (Proceedings of Pacific Graphics 2016)

  [Paper]

• Fast Feature-Oriented Visual Connection for Large Image Collections

  Qingan Yan, Zhan Xu, Chunxia Xiao

  Computer Graphics Forum (CGF), 2014. (Proceedings of Pacific Graphics 2014)

  [Paper]

Journal Reviewer: TPAMI, TIP, TVCG, TMM, TCSVT, TCYB, Neurocomputing, PR, TVC

Conference Reviewer: CVPR, ICCV, ECCV, WACV, ACCV, 3DV, CRV, ICLR, AAAI, EG

Presentations CVPR 2017, Pacific Graphics 2014&2016

Blogs

• Research Recorder
  A research-related memory palace for myself (constructing... subject to... time and mood...).

Tools

• Texture Mapping (C++)
  Our trial implementation of the paper, "Patch-Based Optimization for Image-Based Texture Mapping" SIGGRAPH 2017.

• SfM Viewer (QT & C++)
  The code can be used to display 3D point clouds from .out and .ply files, which are the outputs of bundler_sfm.

• Binocular Measurer (Visual Studio & OpenCV)
  A distance measuring software based on classical binocular stereo algorithm and OpenCV.