Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Taco S. Cohen

Taco S. Cohen

Instance-Adaptive Video Compression: Improving Neural Codecs by Training on the Test Set


Nov 19, 2021
Ties van Rozendaal, Johann Brehmer, Yunfan Zhang, Reza Pourreza, Taco S. Cohen

Figure 1 for Instance-Adaptive Video Compression: Improving Neural Codecs by Training on the Test Set
Figure 2 for Instance-Adaptive Video Compression: Improving Neural Codecs by Training on the Test Set
Figure 3 for Instance-Adaptive Video Compression: Improving Neural Codecs by Training on the Test Set
Figure 4 for Instance-Adaptive Video Compression: Improving Neural Codecs by Training on the Test Set

  Access Paper or Ask Questions

Skip-Convolutions for Efficient Video Processing


Apr 23, 2021
Amirhossein Habibian, Davide Abati, Taco S. Cohen, Babak Ehteshami Bejnordi

Figure 1 for Skip-Convolutions for Efficient Video Processing
Figure 2 for Skip-Convolutions for Efficient Video Processing
Figure 3 for Skip-Convolutions for Efficient Video Processing
Figure 4 for Skip-Convolutions for Efficient Video Processing
* CVPR 2021 
  Access Paper or Ask Questions

A Combined Deep Learning based End-to-End Video Coding Architecture for YUV Color Space


Apr 01, 2021
Ankitesh K. Singh, Hilmi E. Egilmez, Reza Pourreza, Muhammed Coban, Marta Karczewicz, Taco S. Cohen

Figure 1 for A Combined Deep Learning based End-to-End Video Coding Architecture for YUV Color Space
Figure 2 for A Combined Deep Learning based End-to-End Video Coding Architecture for YUV Color Space
Figure 3 for A Combined Deep Learning based End-to-End Video Coding Architecture for YUV Color Space
Figure 4 for A Combined Deep Learning based End-to-End Video Coding Architecture for YUV Color Space
* 5 pages, submitted to as a conference paper. arXiv admin note: text overlap with arXiv:2103.01760 
  Access Paper or Ask Questions

Transform Network Architectures for Deep Learning based End-to-End Image/Video Coding in Subsampled Color Spaces


Feb 27, 2021
Hilmi E. Egilmez, Ankitesh K. Singh, Muhammed Coban, Marta Karczewicz, Yinhao Zhu, Yang Yang, Amir Said, Taco S. Cohen

Figure 1 for Transform Network Architectures for Deep Learning based End-to-End Image/Video Coding in Subsampled Color Spaces
Figure 2 for Transform Network Architectures for Deep Learning based End-to-End Image/Video Coding in Subsampled Color Spaces
Figure 3 for Transform Network Architectures for Deep Learning based End-to-End Image/Video Coding in Subsampled Color Spaces
Figure 4 for Transform Network Architectures for Deep Learning based End-to-End Image/Video Coding in Subsampled Color Spaces
* 10 pages, submitted to an IEEE journal 
  Access Paper or Ask Questions

Overfitting for Fun and Profit: Instance-Adaptive Data Compression


Jan 21, 2021
Ties van Rozendaal, Iris A. M. Huijben, Taco S. Cohen

Figure 1 for Overfitting for Fun and Profit: Instance-Adaptive Data Compression
Figure 2 for Overfitting for Fun and Profit: Instance-Adaptive Data Compression
Figure 3 for Overfitting for Fun and Profit: Instance-Adaptive Data Compression
Figure 4 for Overfitting for Fun and Profit: Instance-Adaptive Data Compression
* Accepted at ICLR 20201 
  Access Paper or Ask Questions

Lossy Compression with Distortion Constrained Optimization


May 08, 2020
Ties van Rozendaal, Guillaume Sautière, Taco S. Cohen

Figure 1 for Lossy Compression with Distortion Constrained Optimization
Figure 2 for Lossy Compression with Distortion Constrained Optimization
Figure 3 for Lossy Compression with Distortion Constrained Optimization
Figure 4 for Lossy Compression with Distortion Constrained Optimization
* Accepted as a CVPR 2020 workshop paper: Workshop and Challenge on Learned Image Compression (CLIC) 
  Access Paper or Ask Questions

A Data and Compute Efficient Design for Limited-Resources Deep Learning


Apr 21, 2020
Mirgahney Mohamed, Gabriele Cesa, Taco S. Cohen, Max Welling

Figure 1 for A Data and Compute Efficient Design for Limited-Resources Deep Learning
Figure 2 for A Data and Compute Efficient Design for Limited-Resources Deep Learning
Figure 3 for A Data and Compute Efficient Design for Limited-Resources Deep Learning
* Accepted for poster presentation at the Practical Machine Learning for Developing Countries (PML4DC) workshop, ICLR 2020 
  Access Paper or Ask Questions

Learning Discrete Distributions by Dequantization


Jan 30, 2020
Emiel Hoogeboom, Taco S. Cohen, Jakub M. Tomczak

Figure 1 for Learning Discrete Distributions by Dequantization
Figure 2 for Learning Discrete Distributions by Dequantization
Figure 3 for Learning Discrete Distributions by Dequantization
Figure 4 for Learning Discrete Distributions by Dequantization

  Access Paper or Ask Questions

Video Compression With Rate-Distortion Autoencoders


Aug 14, 2019
Amirhossein Habibian, Ties van Rozendaal, Jakub M. Tomczak, Taco S. Cohen

Figure 1 for Video Compression With Rate-Distortion Autoencoders
Figure 2 for Video Compression With Rate-Distortion Autoencoders
Figure 3 for Video Compression With Rate-Distortion Autoencoders
Figure 4 for Video Compression With Rate-Distortion Autoencoders
* Accepted to ICCV 2019 
  Access Paper or Ask Questions

Covariance in Physics and Convolutional Neural Networks


Jun 06, 2019
Miranda C. N. Cheng, Vassilis Anagiannis, Maurice Weiler, Pim de Haan, Taco S. Cohen, Max Welling

Figure 1 for Covariance in Physics and Convolutional Neural Networks

  Access Paper or Ask Questions

Gauge Equivariant Convolutional Networks and the Icosahedral CNN


Feb 11, 2019
Taco S. Cohen, Maurice Weiler, Berkay Kicanaoglu, Max Welling

Figure 1 for Gauge Equivariant Convolutional Networks and the Icosahedral CNN
Figure 2 for Gauge Equivariant Convolutional Networks and the Icosahedral CNN

  Access Paper or Ask Questions

Explorations in Homeomorphic Variational Auto-Encoding


Jul 12, 2018
Luca Falorsi, Pim de Haan, Tim R. Davidson, Nicola De Cao, Maurice Weiler, Patrick Forré, Taco S. Cohen

Figure 1 for Explorations in Homeomorphic Variational Auto-Encoding
Figure 2 for Explorations in Homeomorphic Variational Auto-Encoding
Figure 3 for Explorations in Homeomorphic Variational Auto-Encoding
Figure 4 for Explorations in Homeomorphic Variational Auto-Encoding
* 16 pages, 8 figures, ICML workshop on Theoretical Foundations and Applications of Deep Generative Models 
  Access Paper or Ask Questions

Sample Efficient Semantic Segmentation using Rotation Equivariant Convolutional Networks


Jul 02, 2018
Jasper Linmans, Jim Winkens, Bastiaan S. Veeling, Taco S. Cohen, Max Welling

Figure 1 for Sample Efficient Semantic Segmentation using Rotation Equivariant Convolutional Networks
Figure 2 for Sample Efficient Semantic Segmentation using Rotation Equivariant Convolutional Networks
Figure 3 for Sample Efficient Semantic Segmentation using Rotation Equivariant Convolutional Networks
* Presented at the ICML workshop: Towards learning with limited labels: Equivariance, Invariance, and Beyond, 2018 
  Access Paper or Ask Questions

3D G-CNNs for Pulmonary Nodule Detection


Apr 12, 2018
Marysia Winkels, Taco S. Cohen

Figure 1 for 3D G-CNNs for Pulmonary Nodule Detection
Figure 2 for 3D G-CNNs for Pulmonary Nodule Detection
Figure 3 for 3D G-CNNs for Pulmonary Nodule Detection
Figure 4 for 3D G-CNNs for Pulmonary Nodule Detection

  Access Paper or Ask Questions

Intertwiners between Induced Representations (with Applications to the Theory of Equivariant Neural Networks)


Mar 30, 2018
Taco S. Cohen, Mario Geiger, Maurice Weiler

Figure 1 for Intertwiners between Induced Representations (with Applications to the Theory of Equivariant Neural Networks)
Figure 2 for Intertwiners between Induced Representations (with Applications to the Theory of Equivariant Neural Networks)

  Access Paper or Ask Questions

HexaConv


Mar 06, 2018
Emiel Hoogeboom, Jorn W. T. Peters, Taco S. Cohen, Max Welling

Figure 1 for HexaConv
Figure 2 for HexaConv
Figure 3 for HexaConv
Figure 4 for HexaConv

  Access Paper or Ask Questions

Spherical CNNs


Feb 25, 2018
Taco S. Cohen, Mario Geiger, Jonas Koehler, Max Welling

Figure 1 for Spherical CNNs
Figure 2 for Spherical CNNs
Figure 3 for Spherical CNNs
Figure 4 for Spherical CNNs
* Proceedings of the 6th International Conference on Learning Representations (ICLR), 2018 
  Access Paper or Ask Questions

A New Method to Visualize Deep Neural Networks


Jun 12, 2017
Luisa M. Zintgraf, Taco S. Cohen, Max Welling

Figure 1 for A New Method to Visualize Deep Neural Networks
Figure 2 for A New Method to Visualize Deep Neural Networks
Figure 3 for A New Method to Visualize Deep Neural Networks
Figure 4 for A New Method to Visualize Deep Neural Networks
* Please note that this version of the article is outdated. The new version (published at ICLR2017) includes additional experiments on MRI scans and can be found at arXiv:1702.04595 
  Access Paper or Ask Questions

Steerable CNNs


Dec 27, 2016
Taco S. Cohen, Max Welling

Figure 1 for Steerable CNNs
Figure 2 for Steerable CNNs
Figure 3 for Steerable CNNs
Figure 4 for Steerable CNNs

  Access Paper or Ask Questions

Group Equivariant Convolutional Networks


Jun 03, 2016
Taco S. Cohen, Max Welling

Figure 1 for Group Equivariant Convolutional Networks
Figure 2 for Group Equivariant Convolutional Networks
Figure 3 for Group Equivariant Convolutional Networks
Figure 4 for Group Equivariant Convolutional Networks
* Proceedings of the International Conference on Machine Learning (ICML), 2016 
  Access Paper or Ask Questions

Harmonic Exponential Families on Manifolds


May 20, 2015
Taco S. Cohen, Max Welling

Figure 1 for Harmonic Exponential Families on Manifolds
Figure 2 for Harmonic Exponential Families on Manifolds
Figure 3 for Harmonic Exponential Families on Manifolds
* fixed typo 
  Access Paper or Ask Questions

Transformation Properties of Learned Visual Representations


Apr 07, 2015
Taco S. Cohen, Max Welling

Figure 1 for Transformation Properties of Learned Visual Representations
Figure 2 for Transformation Properties of Learned Visual Representations
Figure 3 for Transformation Properties of Learned Visual Representations
Figure 4 for Transformation Properties of Learned Visual Representations
* T.S. Cohen & M. Welling, Transformation Properties of Learned Visual Representations. In International Conference on Learning Representations (ICLR), 2015 
  Access Paper or Ask Questions