...

Anton Obukhov

About

I am a postdoctoral researcher in the Photogrammetry and Remote Sensing group at ETH Zurich, broadly interested in Computer Vision and Machine Learning research. I received my PhD in the Computer Vision Laboratory at ETH Zurich. Even before that, I spent a decade working in the industry: I helped NVIDIA revolutionize parallel computations with NVIDIA CUDA technology and later joined Ubiquiti Networks to build multiple video camera products.

Publications

DGInStyle: Domain-Generalizable Semantic Segmentation with Image Diffusion Models and Stylized Semantic Control

Yuru Jia Lukas Hoyer Shengyu Huang Tianfu Wang Luc Van Gool Konrad Schindler Anton Obukhov

Preprint, under review

Marigold: Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation

Bingxin Ke Anton Obukhov Shengyu Huang Nando Metzger Rodrigo Caye Daudt Konrad Schindler

Preprint, under review

Breathing New Life into 3D Assets with Generative Repainting

Tianfu Wang Menelaos Kanakis Konrad Schindler Luc Van Gool Anton Obukhov

🦄 Oral talk at the British Machine Vision Conference (BMVC) 2023

Point2CAD: Reverse Engineering CAD Models from 3D Point Clouds

Yujia Liu Anton Obukhov Jan Dirk Wegner Konrad Schindler

Preprint, under review

EDAPS: Enhanced Domain-Adaptive Panoptic Segmentation

Suman Saha · Lukas Hoyer · Anton Obukhov · Dengxin Dai · Luc Van Gool

International Conference on Computer Vision (ICCV) 2023

DiffDreamer: Consistent Single-view Perpetual View Generation with Conditional Diffusion Models

Shengqu Cai · Eric Ryan Chan · Songyou Peng · Mohamad Shahbazi · Anton Obukhov · Luc Van Gool · Gordon Wetzstein

International Conference on Computer Vision (ICCV) 2023

TT-NF: Tensor Train Neural Fields

Anton Obukhov · Mikhail Usvyatsov · Christos Sakaridis · Konrad Schindler · Luc Van Gool

Preprint, under review

Towards Practical Control of Singular Values of Convolutional Layers

Alexandra Senderovich · Ekaterina Bulatova · Anton Obukhov · Maxim Rakhuba

Conference on Neural Information Processing Systems (NeurIPS) 2022

Pix2NeRF: Unsupervised Conditional π-GAN for Single Image to Neural Radiance Fields Translation

Shengqu Cai · Anton Obukhov · Dengxin Dai · Luc Van Gool

Conference on Computer Vision and Pattern Recognition (CVPR) 2022

Spectral Tensor Train Parameterization of Deep Learning Layers

Anton Obukhov · Maxim Rakhuba · Alexander Liniger · Zhiwu Huang · Stamatios Georgoulis · Dengxin Dai · Luc Van Gool

International Conference on Artificial Intelligence and Statistics (AISTATS) 2021

Exploring Relational Context for Multi-Task Dense Prediction

David Bruggemann · Menelaos Kanakis · Anton Obukhov · Stamatios Georgoulis · Luc Van Gool

International Conference on Computer Vision (ICCV) 2021

Learning to Relate Depth and Semantics for Unsupervised Domain Adaptation

Suman Saha · Anton Obukhov · Danda Pani Paudel · Menelaos Kanakis · Yuhua Chen · Stamatios Georgoulis · Luc Van Gool

Conference on Computer Vision and Pattern Recognition (CVPR) 2021

Reparameterizing Convolutions for Incremental Multi-Task Learning Without Task Interference

Menelaos Kanakis · David Bruggemann · Suman Saha · Stamatios Georgoulis · Anton Obukhov · Luc Van Gool

European Conference on Computer Vision (ECCV) 2020

T-Basis: a Compact Representation for Neural Networks

Anton Obukhov · Maxim Rakhuba · Stamatios Georgoulis · Menelaos Kanakis · Dengxin Dai · Luc Van Gool

International Conference on Machine Learning (ICML) 2020

Haar Classifiers for Object Detection with CUDA

Anton Obukhov

GPU Computing Gems Emerald Edition 2011

Discrete Cosine Transform for 8x8 Blocks with CUDA

Anton Obukhov · Alexander Kharlamov

NVIDIA CUDA SDK 2008

News

  • 2023.09: One papers was accepted at BMVC 2023 as Oral: “Breathing New Life into 3D Assets with Generative Repainting.”
  • 2023.07: Two papers were accepted at ICCV 2023: “DiffDreamer: Towards Consistent Unsupervised Single-view Scene Extrapolation with Conditional Diffusion Models” and “EDAPS: Enhanced Domain-Adaptive Panoptic Segmentation.”
  • 2023.04: One paper was accepted at CVPR Structural and Compositional Learning on 3D Data 2023 workshop: “Point2CAD: Reverse Engineering CAD Models from 3D Point Clouds.”
  • 2023.03: One paper was accepted at ICLR Neural Fields 2023 workshop: “TT-NF: Tensor Train Neural Fields.”
  • 2022.11: One paper was accepted at SlowDNN 2023 workshop: “TT-NF: Tensor Train Neural Fields.”
  • 2022.11: Gave an invited talk at Huawei Zurich about our latest work, “TT-NF: Tensor Train Neural Fields.”
  • 2022.11: Started as a postdoc in the Photogrammetry and Remote Sensing Group at ETH Zurich with Professor Konrad Schindler.
  • 2022.09: Successfully defended my Ph.D. thesis titled “Tensor Decompositions in Deep Learning.”
  • 2022.08: One paper was accepted at NeurIPS 2022: “Towards Practical Control of Singular Values of Convolutional Layers.”
  • 2022.07: Attended ICVSS summer school 2022.
  • 2022.06: Two patents were published based on the AISTATS 2021 paper “Spectral Tensor Train Parameterization of Deep Learning Layers.”
  • 2022.03: One paper was accepted at CVPR 2022: “Pix2NeRF: Unsupervised Conditional π-GAN for Single Image to Neural Radiance Fields Translation.”
  • 2021.12: One patent was published based on the ICML 2020 paper “T-Basis: a Compact Representation for Neural Networks.”
  • 2021.07: One paper was accepted at ICCV 2021: “Exploring Relational Context for Multi-Task Dense Prediction.”
  • 2021.06: One paper was accepted into Sparsity in Neural Networks 2021 Workshop: “Spectral Tensor Train Parameterization of Deep Learning Layers.”
  • 2021.03: One paper was accepted at CVPR 2021: “Learning to Relate Depth and Semantics for Unsupervised Domain Adaptation.”
  • 2021.04: torch-fidelity will be featured at PyTorch Ecosystem Day 2021.
  • 2021.01: One paper was accepted at AISTATS 2021: “Spectral Tensor Train Parameterization of Deep Learning Layers.”
  • 2020.07: One paper was accepted at ECCV 2020: “Reparameterizing Convolutions for Incremental Multi-task Learning without Task Interference”.
  • 2020.06: One paper was accepted at ICML 2020: “T-Basis: a Compact Representation for Neural Networks.”
  • 2018.05: Started as a Ph.D. candidate in the Computer Vision Lab at ETH Zurich with Professor Luc Van Gool.