Achieving rotational invariance with bessel-convolutional neural networks

V Delchevalerie, A Bibal, B Frénay… - Advances in Neural …, 2021 - proceedings.neurips.cc
For many applications in image analysis, learning models that are invariant to translations
and rotations is paramount. This is the case, for example, in medical imaging where the …

Self-interpretable model with transformation equivariant interpretation

Y Wang, X Wang - Advances in Neural Information …, 2021 - proceedings.neurips.cc
With the proliferation of machine learning applications in the real world, the demand for
explaining machine learning predictions continues to grow especially in high-stakes fields …

Few-shot shape recognition by learning deep shape-aware features

W Shi, C Lu, M Shao, Y Zhang… - Proceedings of the …, 2024 - openaccess.thecvf.com
Traditional shape descriptors have been gradually replaced by convolutional neural
networks due to their superior performance in feature extraction and classification. The state …

Adoption of image-driven machine learning for microstructure characterization and materials design: A perspective

A Baskaran, EJ Kautz, A Chowdhary, W Ma, B Yener… - Jom, 2021 - Springer
The recent surge in the adoption of machine learning techniques for materials design,
discovery, and characterization has resulted in increased interest in and application of …

Breast cancer classification using equivariance transition in group convolutional neural networks

Z Sani, R Prasad, EKM Hashim - IEEE Access, 2023 - ieeexplore.ieee.org
In computer vision, rotation equivariance and translation invariance are properties of a
representation that preserve the geometric structure of a transformed input. These properties …

An imbalance-aware nuclei segmentation methodology for H&E stained histopathology images

E Hancer, M Traore, R Samet, Z Yıldırım… - … Signal Processing and …, 2023 - Elsevier
A key step in computational pathology is to automate the laborious process of manual nuclei
segmentation in Hematoxylin and Eosin (H&E) stained whole slide images (WSIs). Despite …

Practical equivariances via relational conditional neural processes

D Huang, M Haussmann, U Remes… - Advances in …, 2023 - proceedings.neurips.cc
Abstract Conditional Neural Processes (CNPs) are a class of metalearning models popular
for combining the runtime efficiency of amortized inference with reliable uncertainty …