no code implementations • 26 Apr 2024 • Brian B. Moser, Ahmed Anwar, Federico Raue, Stanislav Frolov, Andreas Dengel
This distinction is crucial as it circumvents the need to precisely model real-world degradations, which limits contemporary blind image SR research.
no code implementations • 11 Apr 2024 • Stanislav Frolov, Brian B. Moser, Sebastian Palacio, Andreas Dengel
We present ObjBlur, a novel curriculum learning approach to improve layout-to-image generation models, where the task is to produce realistic images from layouts composed of boxes and labels.
no code implementations • 6 Mar 2024 • Brian B. Moser, Federico Raue, Sebastian Palacio, Stanislav Frolov, Andreas Dengel
In response to these limitations, the concept of distilling the information on a dataset into a condensed set of (synthetic) samples, namely a distilled dataset, emerged.
no code implementations • 1 Jan 2024 • Brian B. Moser, Arundhati S. Shanbhag, Federico Raue, Stanislav Frolov, Sebastian Palacio, Andreas Dengel
Diffusion Models (DMs) have disrupted the image Super-Resolution (SR) field and further closed the gap between image quality and human perceptual preferences.
no code implementations • 15 Aug 2023 • Brian B. Moser, Stanislav Frolov, Federico Raue, Sebastian Palacio, Andreas Dengel
To address this, we introduce "You Only Diffuse Areas" (YODA), a dynamic attention-guided diffusion method for image SR. YODA selectively focuses on spatial regions using attention maps derived from the low-resolution image and the current time step in the diffusion process.
1 code implementation • 10 Jul 2023 • Brian B. Moser, Stanislav Frolov, Federico Raue, Sebastian Palacio, Andreas Dengel
This work introduces Differential Wavelet Amplifier (DWA), a drop-in module for wavelet-based image Super-Resolution (SR).
no code implementations • 10 Apr 2023 • João Maria Janeiro, Stanislav Frolov, Alaaeldin El-Nouby, Jakob Verbeek
For example, for segmentation mIoU is reduced from 44. 5 to 30. 5 mIoU when compressing to 0. 1 bpp using the best compression model we evaluated.
1 code implementation • 4 Apr 2023 • Brian Moser, Stanislav Frolov, Federico Raue, Sebastian Palacio, Andreas Dengel
This paper presents a novel Diffusion-Wavelet (DiWa) approach for Single-Image Super-Resolution (SISR).
no code implementations • 27 Sep 2022 • Brian Moser, Federico Raue, Stanislav Frolov, Jörn Hees, Sebastian Palacio, Andreas Dengel
With the advent of Deep Learning (DL), Super-Resolution (SR) has also become a thriving research area.
no code implementations • 5 Apr 2022 • Stanislav Frolov, Prateek Bansal, Jörn Hees, Andreas Dengel
Our results demonstrate the capability of our approach to generate plausible images of complex scenes using region captions.
no code implementations • 21 May 2021 • Ricard Durall, Stanislav Frolov, Jörn Hees, Federico Raue, Franz-Josef Pfreundt, Andreas Dengel, Janis Keupe
Transformer models have recently attracted much interest from computer vision researchers and have since been successfully employed for several problems traditionally addressed with convolutional neural networks.
1 code implementation • 25 Mar 2021 • Stanislav Frolov, Avneesh Sharma, Jörn Hees, Tushar Karayil, Federico Raue, Andreas Dengel
In this paper, we propose a method for attribute controlled image synthesis from layout which allows to specify the appearance of individual objects without affecting the rest of the image.
no code implementations • 25 Jan 2021 • Stanislav Frolov, Tobias Hinz, Federico Raue, Jörn Hees, Andreas Dengel
With the advent of generative adversarial networks, synthesizing images from textual descriptions has recently become an active research area.
no code implementations • LANTERN (COLING) 2020 • Stanislav Frolov, Shailza Jolly, Jörn Hees, Andreas Dengel
We create additional training samples by concatenating question and answer (QA) pairs and employ a standard VQA model to provide the T2I model with an auxiliary learning signal.
1 code implementation • 10 Oct 2020 • Fatemeh Azimi, Stanislav Frolov, Federico Raue, Joern Hees, Andreas Dengel
In this work, we study an RNN-based architecture and address some of these issues by proposing a hybrid sequence-to-sequence architecture named HS2S, utilizing a dual mask propagation strategy that allows incorporating the information obtained from correspondence matching.