Home | Research | Groups | Reinhard Heckel

Research Group Reinhard Heckel

Link to Reinhard Heckel

Reinhard Heckel

Prof. Dr.

Principal Investigator

Machine Learning

Reinhard Heckel

is Professor for Machine Learning at TU Munich.

His research is centered on machine learning and information processing. He focuses on developing algorithms and theoretical foundations for deep learning, particularly in medical imaging application, and on establishing mathematical and empirical underpinnings for machine learning. Additionally, he works on DNA data storage and the utilization of DNA as a digital information technology.

Team members @MCML

Link to Florian Fürnrohr

Florian Fürnrohr

Machine Learning

Publications @MCML

[4]
Y. Mansour, X. Zhong, S. Caglar and R. Heckel.
TTT-MIM: Test-Time Training with Masked Image Modeling for Denoising Distribution Shifts.
18th European Conference on Computer Vision (ECCV 2024). Milano, Italy, Sep 29-Oct 04, 2024. To be published.
MCML Authors
Link to Reinhard Heckel

Reinhard Heckel

Prof. Dr.

Machine Learning


[3]
Y. Mansour and R. Heckel.
GAMA-IR: Global Additive Multidimensional Averaging for Fast Image Restoration.
Preprint at arXiv (Apr. 2024). arXiv.
Abstract

Deep learning-based methods have shown remarkable success for various image restoration tasks such as denoising and deblurring. The current state-of-the-art networks are relatively deep and utilize (variants of) self attention mechanisms. Those networks are significantly slower than shallow convolutional networks, which however perform worse. In this paper, we introduce an image restoration network that is both fast and yields excellent image quality. The network is designed to minimize the latency and memory consumption when executed on a standard GPU, while maintaining state-of-the-art performance. The network is a simple shallow network with an efficient block that implements global additive multidimensional averaging operations. This block can capture global information and enable a large receptive field even when used in shallow networks with minimal computational overhead. Through extensive experiments and evaluations on diverse tasks, we demonstrate that our network achieves comparable or even superior results to existing state-of-the-art image restoration networks with less latency. For instance, we exceed the state-of-the-art result on real-world SIDD denoising by 0.11dB, while being 2 to 10 times faster.

MCML Authors
Link to Reinhard Heckel

Reinhard Heckel

Prof. Dr.

Machine Learning


[2]
S. Bamberger, R. Heckel and F. Krahmer.
Approximating Positive Homogeneous Functions with Scale Invariant Neural Networks.
Preprint at arXiv (Aug. 2023). arXiv.
MCML Authors
Link to Reinhard Heckel

Reinhard Heckel

Prof. Dr.

Machine Learning

Link to Felix Krahmer

Felix Krahmer

Prof. Dr.

Optimization & Data Analysis


[1]
Y. Mansour and R. Heckel.
Zero-Shot Noise2Noise: Efficient Image Denoising without any Data.
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2023). Vancouver, Canada, Jun 18-23, 2023. DOI.
MCML Authors
Link to Reinhard Heckel

Reinhard Heckel

Prof. Dr.

Machine Learning