Dr. Daniyar Turmukhambetov
I’m a staff research scientist at Niantic, creators of Pokemon GO.
I’ve joined Niantic through acquisition of Matrix Mill, which I founded together with Dr. Gabriel Brostow and Dr. Michael Firman.
Previously, I was a post-doctoral research associate with Dr. Gabriel Brostow at University College London. I worked on on EU-FP7 project “Capture-Reconstruct-Play” and “Nature-Smart Cities - Sensing Nature in the Queen Elizabeth Olympic Park” EPSRC Impact project. See news article about Nature-Smart Cities project on BBC website.
I completed my PhD in the Department of Computer Science at University College London under supervision of Prof. Jan Kautz and Dr. Simon Prince in collaboration with Dr. Neill D.F. Campbell. My PhD thesis, titled “Synthesizing and Editing Photo-realistic Visual Objects”, explores image synthesis by combining computer graphics techniques with probabilistic generative models of computer vision. During my PhD I did a summer research internship at Creative Technologies Lab of Adobe in Seattle with Dr. Dan B Goldman.
Before starting my PhD, I finished MSc in Computer Graphics, Vision and Imaging at UCL with distinction. My Master Thesis “Intrinsic Images from Color and Depth”, supervised by Dr. Gabriel Brostow, has won “BBC Best Project Prize”. I completed Bachelor of Engineering in “Computer Science and Technology” at Tsinghua University with a scholarship from China National Petroleum Corporation.
Publications
Single Image Depth Prediction with Wavelet Decomposition
Michaël Ramamonjisoa, Michael Firman, Jamie Watson, Vincent Lepetit and Daniyar Turmukhambetov
In Conference on Computer Vision and Pattern Recognition, 2021
project webpage
Learning to Predict Repeatability of Interest Points
Anh-Dzung Doan, Daniyar Turmukhambetov, Yasir Latif, Tat-Jun Chin and Soohyun Bae
In International Conference on Robotics and Automation, 2021
project webpage
Learning Stereo from Single Images
Jamie Watson, Oisin Mac Aodha, Daniyar Turmukhambetov, Gabriel J. Brostow and Michael Firman
In European Conference on Computer Vision, 2020
project webpage
Single-Image Depth Prediction Makes Features Matching Easier
Carl Toft, Daniyar Turmukhambetov, Tosten Sattler, Fredrik Kahl and Gabriel J. Brostow
In European Conference on Computer Vision, 2020
project webpage
Predicting Visual Overlap of Images Through Interpretable Non-Metric Box Embeddings
Anita Rau, Guillermo Garcia-Hernando, Danail Stoyanov, Gabriel J. Brostow and Daniyar Turmukhambetov
In European Conference on Computer Vision, 2020
project webpage
Image Stylization for Robust Features
Iaroslav Melekhov, Gabriel J. Brostow, Juho Kannala, Daniyar Turmukhambetov
arxiv
Self-Supervised Monocular Depth Hints
Jamie Watson, Michael Firman, Gabriel J. Brostow and Daniyar Turmukhambetov
In International Conference on Computer Vision, 2019
project webpage
HILC: Domain-Independent PbD System Via Computer Vision and Follow-Up Questions
Thanapong Intharah, Daniyar Turmukhambetov and Gabriel J. Brostow
In ACM Transactions on Interactive Intelligent Systems (TiiS), 2019
preprint
Interpretable Transformations with Encoder-Decoder Networks
Daniel E. Worrall, Stephan J. Garbin, Daniyar Turmukhambetov and Gabriel J. Brostow
In International Conference on Computer Vision, 2017
project webpage
Harmonic Networks: Deep Translation and Rotation Equivariance
Daniel E. Worrall, Stephan J. Garbin, Daniyar Turmukhambetov and Gabriel J. Brostow
In Conference on Computer Vision and Pattern Recognition, 2017
project webpage
Help, It Looks Confusing: GUI Task Automation Through Demonstration and Follow-up Questions
Thanapong Intharah, Daniyar Turmukhambetov and Gabriel J. Brostow
In ACM Intelligent User Interfaces (IUI), 2017
Best Student Paper Honorable Mention
project webpage
Synthesizing and Editing Photo-realistic Visual Objects
Daniyar Turmukhambetov
PhD Thesis at University College London, 2016
UCL Thesis Repository
Modeling Object Appearance using Context-Conditioned Component Analysis
Daniyar Turmukhambetov, Neill D.F. Campbell, Simon J.D. Prince and Jan Kautz
In Conference on Computer Vision and Pattern Recognition, 2015
project webpage
Interactive Sketch-Driven Image Synthesis
Daniyar Turmukhambetov, Neill D.F. Campbell, Dan B Goldman and Jan Kautz
In Computer Graphics Forum, 2015
project webpage
Intrinsic Images from Color and Depth
Daniyar Turmukhambetov
MSc Thesis at University College London, 2011