• English
  • Deutsch
  • Log In
    Password Login
    Research Outputs
    Fundings & Projects
    Researchers
    Institutes
    Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Konferenzschrift
  4. Improving Feature Stability during Upsampling - Spectral Artifacts and the Importance of Spatial Context
 
  • Details
  • Full
Options
2025
Conference Paper
Title

Improving Feature Stability during Upsampling - Spectral Artifacts and the Importance of Spatial Context

Abstract
Pixel-wise predictions are required in a wide variety of tasks such as image restoration, image segmentation, or disparity estimation. Common mod els involve several stages of data resampling, in which the resolution of feature maps is first reduced to aggregate information and then increased to generate a high-resolution output. Previous works have shown that resampling operations are subject to artifacts such as aliasing. During downsampling, aliases have been shown to compromise the prediction stability of image classifiers. During up sampling, they have been leveraged to detect generated content. Yet, the effect of aliases during upsampling has not yet been discussed w.r.t. the stability and ro bustness of pixel-wise predictions. While falling under the same term (aliasing), the challenges for correct upsampling in neural networks differ significantly from those during downsampling: when downsampling, some high frequencies can not be correctly represented and have to be removed to avoid aliases. However, when upsampling for pixel-wise predictions, we actually require the model to restore such high frequencies that can not be encoded in lower resolutions. The applica tion of findings from signal processing is therefore a necessary but not a sufficient condition to achieve the desirable output. In contrast, we find that the availability of large spatial context during upsampling allows to provide stable, high-quality pixel-wise predictions, even when fully learning all filter weights.
Author(s)
Agnihotri, Shashank  
Grabinski, Julia
Fraunhofer-Institut für Techno- und Wirtschaftsmathematik ITWM  
Keuper, Margret  
Mainwork
Computer Vision - ECCV 2024. Proceedings. Part LVIII  
Conference
European Conference on Computer Vision 2024  
DOI
10.1007/978-3-031-73636-0_21
Language
English
Fraunhofer-Institut für Techno- und Wirtschaftsmathematik ITWM  
Keyword(s)
  • Pixel-wise predictions

  • image restoration

  • image segmentation

  • Cookie settings
  • Imprint
  • Privacy policy
  • Api
  • Contact
© 2024