ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2101.05846
27
6

How Shift Equivariance Impacts Metric Learning for Instance Segmentation

14 January 2021
J. L. Rumberger
Xiaoyan Yu
Peter Hirsch
Melanie Dohmen
Vanessa Emanuela Guarino
Ashkan Mokarian
Lisa Mais
Jan Funke
Dagmar Kainmueller
ArXivPDFHTML
Abstract

Metric learning has received conflicting assessments concerning its suitability for solving instance segmentation tasks. It has been dismissed as theoretically flawed due to the shift equivariance of the employed CNNs and their respective inability to distinguish same-looking objects. Yet it has been shown to yield state of the art results for a variety of tasks, and practical issues have mainly been reported in the context of tile-and-stitch approaches, where discontinuities at tile boundaries have been observed. To date, neither of the reported issues have undergone thorough formal analysis. In our work, we contribute a comprehensive formal analysis of the shift equivariance properties of encoder-decoder-style CNNs, which yields a clear picture of what can and cannot be achieved with metric learning in the face of same-looking objects. In particular, we prove that a standard encoder-decoder network that takes ddd-dimensional images as input, with lll pooling layers and pooling factor fff, has the capacity to distinguish at most fdlf^{dl}fdl same-looking objects, and we show that this upper limit can be reached. Furthermore, we show that to avoid discontinuities in a tile-and-stitch approach, assuming standard batch size 1, it is necessary to employ valid convolutions in combination with a training output window size strictly greater than flf^lfl, while at test-time it is necessary to crop tiles to size n⋅fln\cdot f^ln⋅fl before stitching, with n≥1n\geq 1n≥1. We complement these theoretical findings by discussing a number of insightful special cases for which we show empirical results on synthetic data.

View on arXiv
Comments on this paper