Michael Tschannen
![](https://mitscha.github.io/assets/img/michael.jpg)
I’m a Research Scientist at Google DeepMind Zurich (formerly Google Brain) broadly interested in multimodal learning for understanding and generation tasks.
Before that I was working on computer vision R&D at Apple Zurich for two years, and spent a year as a postdoc at Google Research Zurich (Brain Team) exploring topics in unsupervised representation learning, generative models, and neural compression. I completed my PhD at ETH Zurich under the supervision of Helmut Bölcskei in late 2018. Prior to that I obtained a MSc (with distinction) from ETH Zurich and a BSc from EPFL, both in Electrical Engineering and Information Technology. In fall 2017, I interned at Amazon AI in Palo Alto, CA, and in fall 2018 I was a part-time research consultant working with Google Research Zurich (Brain Team).
Contact: mi.<last name><at>gmail.com
News
May 4, 2024 | Check out recent code releases for GIVT (link) and CapPa (link), and recent talks on CLIPPO (link) and CapPa (link). |
Aug 15, 2022 | I re-joined Google. |
Oct 10, 2020 | HiFiC brings generative image compression to the next level! Check out the demo page and the Hacker News Thread. |
Mar 24, 2020 | Two papers accepted for presentation at CVPR 2020! |
Jan 25, 2020 | I’m happy to announce that I obtained the ETH Medal (outstanding thesis award) for my PhD thesis! |
Publications
*denotes equal contribution. See Google Scholar for a potentially more up-to-date list.
2024
-
PaLI-X: On scaling up a multilingual vision and language model In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024
-
Finite scalar quantization: VQ-VAE made simple In Proc. International Conference on Learning Representations (ICLR), 2024 [colab]
-
Towards truly zero-shot compositional visual reasoning with LLMs as programmers Transactions on Machine Learning Research (TMLR), 2024
2023
-
Image captioners are scalable vision learners too In Advances in Neural Information Processing Systems (NeurIPS), 2023 oral presentation [talk] [code]
-
M2T: Masking transformers twice for faster decoding In Proc. IEEE International Conference on Computer Vision (ICCV), 2023
-
Scaling vision transformers to 22 billion parameters In Proc. International Conference on Machine Learning (ICML), 2023
-
CLIPPO: Image-and-Language Understanding from Pixels Only In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023 [code] [colab] [talk]
-
FlexiViT: One Model for All Patch Sizes In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023 [code]
2022
-
Neural Face Video Compression using Multiple Views In Proc. IEEE Conf. on Computer Vision and Pattern Recognition Workshops (CVPRW), 2022 Workshop and Challenge on Learned Image Compression (CLIC) Best Student Paper Award
2021
-
On Robustness and Transferability of Convolutional Neural Networks In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021
-
Representation learning from videos in-the-wild: An object-centric approach In Proc. IEEE Winter Conference on Applications of Computer Vision (WACV), 2021
2020
-
High-Fidelity Generative Image Compression In Advances in Neural Information Processing Systems (NeurIPS), 2020 oral presentation
-
Automatic shortcut removal for self-supervised representation learning In Proc. International Conference on Machine Learning (ICML), 2020
-
Weakly-supervised disentanglement without compromises In Proc. International Conference on Machine Learning (ICML), 2020
-
Self-supervised learning of video-induced visual invariances In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
-
Learning better lossless image compression using lossy compression In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
-
On mutual information maximization for representation learning In Proc. International Conference on Learning Representations (ICLR), 2020
-
Disentangling factors of variation using few labels In Proc. International Conference on Learning Representations (ICLR), 2020
2019
-
Generative adversarial networks for extreme learned image compression In Proc. IEEE International Conference on Computer Vision (ICCV), 2019
-
Practical full resolution learned lossless image compression In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019 oral presentation
-
High-fidelity image generation with fewer labels In Proc. International Conference on Machine Learning (ICML), 2019
2018
-
Deep generative models for distribution-preserving lossy compression In Advances in Neural Information Processing Systems (NeurIPS), 2018
-
Recent advances in autoencoder-based representation learning Bayesian Deep Learning Workshop at NeurIPS 2018, 2018
-
StrassenNets: Deep learning with a multiplication budget In Proc. International Conference on Machine Learning (ICML), 2018 long oral presentation
-
Conditional probability models for deep image compression In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018
-
Towards image understanding from deep compression without decoding In Proc. International Conference on Learning Representations (ICLR), 2018
-
Unsupervised learning: Model-based clustering and learned compression PhD thesis, ETH Zurich, 2018 ETH Medal (outstanding thesis award)
2017
-
Robust nonparametric nearest neighbor random process clustering IEEE Transactions on Signal Processing, 2017
-
A unified optimization view on generalized matching pursuit and Frank-Wolfe In Proc. International Conference on Artificial Intelligence and Statistics (AISTATS), 2017
-
Greedy algorithms for cone constrained optimization with convergence guarantees In Advances in Neural Information Processing Systems (NIPS), 2017
-
Soft-to-hard vector quantization for end-to-end learning compressible representations In Advances in Neural Information Processing Systems (NIPS), 2017
-
Convolutional recurrent neural networks for electrocardiogram classification In Proc. Computing in Cardiology (CinC), 2017 5th place in the PhysioNet/CinC Challenge 2017
-
Deep structured features for semantic segmentation In Proc. European Signal Processing Conference (EUSIPCO), 2017
2016
-
Discrete deep feature extraction: A theory and new architectures In Proc. International Conference on Machine Learning (ICML), 2016
-
Heart sound classification using deep structured features In Proc. Computing in Cardiology (CinC), 2016
-
Regression forest-based automatic estimation of the articular margin plane for shoulder prosthesis planning Medical Image Analysis, 2016
2015
-
Nonparametric nearest neighbor random process clustering In Proc. IEEE International Symposium on Information Theory (ISIT), 2015
2014
-
Subspace clustering of dimensionality-reduced data In Proc. IEEE International Symposium on Information Theory (ISIT), 2014
-
Dimensionality reduction for sparse subspace clustering MS thesis, ETH Zurich, 2014 ETH Medal (outstanding thesis award) and the SEW Eurodrive Foundation Graduate Award
2013
-
A learning-based approach for fast and robust vessel tracking in long ultrasound sequences In Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2013