Efficient Learning of Convolution Weights as Gaussian Mixture Model Posteriors

Lifan Liang
1/30/2024
cs.CV

No confident code implementation found for this paper.

Abstract

In this paper, we showed that the feature map of a convolution layer is equivalent to the unnormalized log posterior of a special kind of Gaussian mixture for image modeling. Then we expanded the model to drive diverse features and proposed a corresponding EM algorithm to learn the model. Learning convolution weights using this approach is efficient, guaranteed to converge, and does not need supervised information. Code is available at: https://github.com/LifanLiang/CALM.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

An Adaptive Orthogonal Convolution Scheme for Efficient and Flexible CNN Architectures

81% match2026

KernelWarehouse: Rethinking the Design of Dynamic Convolution

80% match1 repo1022026

Flyweight FLIM Networks for Salient Object Detection in Biomedical Images

79% match2025

Inducing Semi-Structured Sparsity by Masking for Efficient Model Inference in Convolutional Networks

79% match2024

Deep Gaussian mixture model for unsupervised image segmentation

78% match1 repo72024

Generalized Gaussian Model for Learned Image Compression

78% match2024

Cite this paper

@article{liang2024efficient,
  title  = {Efficient Learning of Convolution Weights as Gaussian Mixture Model Posteriors},
  author = {Lifan Liang},
  year   = {2024},
  eprint = {2401.17400},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2401.17400v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2401.17400.svg)](https://codeofpaper.com/paper/2401.17400)

Discussion