Research map / Detection and segmentation / CNN features
How Shift Equivariance Impacts Metric Learning for Instance Segmentation
Josef Lorenz Rumberger, Xiaoyan Yu, Peter Hirsch, Melanie Dohmen, Vanessa Emanuela Guarino, Ashkan Mokarian, Lisa Mais, Jan Funke, Dagmar Kainmüller
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| train with output window size greater than f^l and crop inference tiles to n*f^l before stitching (ablation) | U-Net with discriminative loss · BBBC006, DSB2018, nuclei3d | false splits and AP 0.5 | Improved |
| add slight Gaussian noise or small elastic deformations to input (ablation) | U-Net on periodic synthetic disks at spacing f^l · synthetic periodic disks | instance distinction | Improved |
| add noise or deformation when tiles are stitched incorrectly (ablation) | U-Net tile-and-stitch · synthetic periodic disks | stitching inconsistencies | No clear effect |
Closest papers by meaning
- Divide and Conquer the Embedding Space for Metric Learning (CVPR 2019 · similarity 0.88)
- Improving Deep Metric Learning by Divide and Conquer (TPAMI 2021 · similarity 0.86)
- Equivariance versus Augmentation for Spherical Images (ICML 2022 · similarity 0.86)
- Learning Equivariant Segmentation with Instance-Unique Querying (NeurIPS 2022 · similarity 0.86)
- Probing Equivariance and Symmetry Breaking in Convolutional Networks (NeurIPS 2025 · similarity 0.85)
- Training-Free Dataset Pruning for Instance Segmentation (ICLR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
