Duolikun Danier

达尼尔·多里坤

prof_pic.jpg

I am currently a Senior Machine Learning Engineer at Sony PlayStation. Previously, I was fortunate to do my postdoc with Dr. Oisin Mac Aodha, Dr. Hakan Bilen, and Dr. Changjian Li in the School of Informatics at the University of Edinburgh, where I worked on video generation, 3D vision, and robotics.

I did my PhD with the Visual Information Laboratory at the University of Bristol, supervised by Prof. David Bull and Dr. Fan Zhang. The focus of my PhD was on low-level computer vision topics such as video frame interpolation, video compression and video quality assessment. Prior to that, I obtained MSc (Distinction) in Machine Learning from UCL under the supervision of Prof. Gabriel Brostow, and obtained BEng (1st Class Hons.) in Electronic and Electrical Engineering also from UCL, supervised by Prof. Yiannis Andreopoulos.

news

May 18, 2026 Recognized as an outstanding reviewer for CVPR 2026.
Oct 01, 2025 Recognized as an outstanding reviewer for ICCV 2025.
Feb 26, 2025 DepthCues is accepted to CVPR 2025.
Feb 24, 2025 I gave a guest lecture for the Machine Learning Practical course at Edinburgh.
Dec 10, 2024 I gave a talk on DepthCues at VisualAI group, Oxford.
Oct 29, 2024 One co-authored paper accepted to WACV 2025.
Oct 22, 2024 Glad to announce that I have successfully defended my PhD thesis. Thank you Prof. Dima Damen and Prof. Alan Bovik for being my examiners!
Sep 24, 2024 Happy to be recognized as an outstanding reviewer for ECCV 2024.
May 21, 2024 Started working as a postdoc with Dr. Oisin Mac Aodha, Dr. Hakan Bilen, and Dr. Changjian Li at the University of Edinburgh.
Dec 09, 2023 The LDMVFI paper is accepted by AAAI 2024.
Oct 24, 2023 One co-authored paper RankDVQA is accepted by WACV 2024.
Oct 21, 2023 The BVI-VFI paper is accepted by IEEE Transaction on Image Processing (TIP).
May 12, 2023 Started a research internship at MediaTek Research.
Dec 11, 2022 The FloLPIPS paper is selected among the best paper finalists at PCS 2022.
Jun 26, 2022 Two papers accepted at ICIP 2022.
Mar 02, 2022 The ST-MFNet paper is accepted by CVPR 2022.
Nov 02, 2020 Started my PhD study at University of Bristol.

selected publications

  1. ECCV
    cdps.jpg
    Training-free Discriminative Patch Mining for Robust Few-Shot Recognition with CLIP
    Zhenzhang Ye, Duolikun Danier, Bo Zhao, and Hakan Bilen
    ECCV, 2026
  2. CVPR Findings
    moonseg3r.png
    MoonSeg3R: Monocular Online Zero-Shot Segment Anything in 3D with Reconstructive Foundation Priors
    Zhipeng Du, Duolikun Danier, Jan Eric Lenssen, and Hakan Bilen
    CVPR Findings, 2026
    arXiv Paper
  3. ICRA
    temporal_trap.gif
    The Temporal Trap: Entanglement in Pre-Trained Visual Representations for Visuomotor Policy Learning
    Nikolaos Tsagkas, Andreas Sochopoulos, Duolikun Danier, Chris Xiaoxuan Lu, and Oisin Mac Aodha
    ICRA, 2026
    arXiv Code Website
  4. IROS
    afa.gif
    Attentive Feature Aggregation or: How Policies Learn to Stop Worrying about Robustness and Attend to Task-Relevant Visual Cues
    Nikolaos Tsagkas, Andreas Sochopoulos, Duolikun Danier, Sethu Vijayakumar, Alexandros Kouris, Oisin Mac Aodha, and 1 more author
    IROS, 2026
    arXiv Code Website
  5. ICMR
    efficientsam3_full.svg
    SAM3-LiteText: An Anatomical Study of the SAM3 Text Encoder for Efficient Vision-Language Segmentation
    Chengxi Zeng, Yuxuan Jiang, Ge Gao, Shuai Wang, Duolikun Danier, Bin Zhu, and 3 more authors
    ICMR, 2026
    arXiv Code Paper Website
  6. arXiv
    vicodr_crop.gif
    View-Consistent Diffusion Representations for 3D-Consistent Video Generation
    Duolikun Danier, Ge Gao, Steven McDonagh, Changjian Li, Hakan Bilen, and Oisin Mac Aodha
    2025
    arXiv Paper Website
  7. CVPR
    depthcues.png
    DepthCues: Evaluating Monocular Depth Perception in Large Vision Models
    Duolikun Danier, Mehmet Aygün, Changjian Li, Hakan Bilen, and Oisin Mac Aodha
    CVPR, 2025
    arXiv Code Data Paper Website
  8. CVPR Workshop
    pvrobo.gif
    When Pre-trained Visual Representations Fall Short: Limitations in Visuo-Motor Robot Learning
    Nikolaos Tsagkas, Andreas Sochopoulos, Duolikun Danier, Chris Xiaoxuan Lu, and Oisin Mac Aodha
    CVPR Embodied AI Workshop (Oral), 2025
    arXiv Paper Website
  9. WACV
    mvad.png
    MVAD: A Multiple Visual Artifact Detector for Video Streaming
    Chen Feng, Duolikun Danier, Fan Zhang, and David Bull
    WACV, 2025
    arXiv Paper Website
  10. AAAI
    ldmvfi_compressed.gif
    LDMVFI: Video Frame Interpolation with Latent Diffusion Models
    Duolikun Danier, Fan Zhang, and David Bull
    AAAI, 2024
    arXiv Code Paper
  11. TIP
    bvivfip2.PNG
    BVI-VFI: A Video Quality Database for Video Frame Interpolation
    Duolikun Danier, Fan Zhang, and David Bull
    IEEE Transactions on Image Processing, 2023
    arXiv Data Paper
  12. CVPR
    stmfnet_compressed.gif
    ST-MFNet: A Spatio-Temporal Multi-Flow Network for Frame Interpolation
    Duolikun Danier, Fan Zhang, and David Bull
    CVPR, 2022
    arXiv Code Paper
  13. WACV
    rankdvqa.PNG
    RankDVQA: Deep VQA based on Ranking-inspired Hybrid Training
    Chen Feng, Duolikun Danier, Fan Zhang, and David Bull
    WACV, 2024
    arXiv Code Paper
  14. NeurIPS Workshop
    deissn.png
    Score Normalization for a Faster Diffusion Exponential Integrator Sampler
    Guoxuan Xia, Duolikun Danier, Ayan Das, Stathi Fotiadis, Farhang Nabiei, Ushnish Sengupta, and 1 more author
    NeurIPS Workshop on Diffusion Models, 2023
    arXiv Code Paper
  15. PCS
    flolpips.PNG
    FloLPIPS: A Bespoke Video Quality Metric for Frame Interpoation
    Duolikun Danier, Fan Zhang, and David Bull
    IEEE Picture Coding Symposium, 2022
    Award arXiv Code Paper
  16. ICIP
    edc.gif
    Enhancing Deformable Convolution based Video Frame Interpolation with Coarse-to-fine 3D CNN
    Duolikun Danier, Fan Zhang, and David Bull
    ICIP, 2022
    arXiv Code Paper