avatar

Ta-Ying Cheng

Research Scientist
Netflix
tayingc [at] netflix.com


About Me

I am currently a research scientist at Netflix, working on the intersection of VLM and diffusion models. I obtained my D.Phil. in Computer Science at the University of Oxford, co-supervised by Profs. Niki Trigoni and Andrew Markham. I am the recipient of the ACE-OPS grant and St. Catherine’s College Overseas Scholarship.

I also had the privilege of collaborating closely with Matheus Gadelha, Thibault Groueix, Matthew Fisher, Soren Pirk, and Radomir Mech during my internships at Adobe Research, and Varun Jampani, Prafull Sharma, and Mark Boss during my internship at Stability AI. Prior to my D.Phil., I worked as a research assistant in Computer Vision Lab, Academia Sinica, supervised by Profs. Tyng-Luh Liu and Hwann-Tzong Chen.

Research Interests

My research focuses on building both assistive and agentic tools for complex graphic design tasks.

During my PhD, I mainly worked on creating enhancing generative models with intuitive 3D-aware controls without losing any of its generalizability for creative content generation, essentially making them into flexible renderers. Some works include controlling camera/lighting (Continuous 3D Words) and materials (ZeST, MARBLE).

More recently, my work has extended to VLMs, with a general goal of creating high-level controls over generative models that require world understanding and complex reasoning on object interactions (VOID).

News

Selected Publications (Full list on Google Scholar)

  1. CVPR
    MARBLE: Material Recomposition and Blending in CLIP-Space
    Ta-Ying Cheng, Prafull Sharma, Mark Boss, Varun Jampani
    Computer Vision and Pattern Recognition (CVPR), 2025.
    Paper Project Website Code

    ECCV
    ZeST: Zero-Shot Material Transfer from a Single Image
    Ta-Ying Cheng, Prafull Sharma, Andrew Markham, Niki Trigoni, Varun Jampani
    European Conference on Computer Vision (ECCV), 2024.
    Paper Project Website Code

    ArXiv
    Gen4Gen: Generative Data Pipeline for Generative Multi-Concept Composition
    Chun-Hsiao Yeh*, Ta-Ying Cheng*, He-Yen Hsieh*, Chuan-En Lin, Yi Ma, Andrew Markham, Niki Trigoni, H.T. Kung, Yubei Chen (*=Equal Contribution)
    ArXiv, 2024.
    Paper Project Website Code

    CVPR
    Learning Continuous 3D Words for Text-to-Image Generation
    Ta-Ying Cheng, Matheus Gadelha, Thibault Groueix, Matthew Fisher, Radomir Mech, Andrew Markham, Niki Trigoni
    Computer Vision and Pattern Recognition (CVPR), 2024.
    Paper Project Website Code

    ICCV
    3DMiner: Discovering Shapes from Large-Scale Unannotated Image Datasets
    Ta-Ying Cheng, Matheus Gadelha, Soren Pirk, Thibault Groueix, Radomir Mech, Andrew Markham, Niki Trigoni
    International Conference on Computer Vision (ICCV), 2023.
    Paper Project Website

    ECCV
    Meta-Sampler: Almost-Univsersal yet Task-Oriented Sampling for Point Clouds
    Ta-Ying Cheng, Qingyong Hu, Qian Xie, Niki Trigoni, Andrew Markham
    European Conference on Computer Vision (ECCV), 2022.
    Paper Code

    AAAI
    Pose Adaptive Dual Mixup for Few-Shot Single-View 3D Reconstruction
    Ta-Ying Cheng*, Hsuan-Ru Yang*, Niki Trigoni, Hwann-Tzong Chen, Tyng-Luh Liu (*=Equal Contribution)
    AAAI Conference on Artifical Intelligence, 2022.
    Paper Code

    CHI
    ARchitect: Building Interactive Virtual Experiences from Physical Affordances by Bringing Human-in-the-Loop
    Chuan-En Lin*, Ta-Ying Cheng*, Xiaojuan Ma (*=Equal Contribution)
    CHI Conference on Human Factors in Computing Systems, 2020.
    Paper Project Website

Services

Conference Reviewers


Powered by Jekyll and Minimal Light theme.