Convolutional Transform learning

Abstract : This work proposes a new representation learning technique called convolutional transform learning. In standard transform learning, a dense basis is learned that analyses the image to generate the representation from the image. Here, we learn a set of independent convolutional filters that operate on the images to produce representations (one corresponding to each filter). The major advantage of our proposed approach is that it is completely unsupervised; unlike CNNs where labeled images are required for training. Moreover, it relies on a well-sounded minimization technique with established convergence guarantees. We have compared the proposed method with dictionary learning and transform learning on standard image classification datasets. Results show that our method improves over the rest by a considerable margin.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [20 references]  Display  Hide  Download
Contributor : Emilie Chouzenoux <>
Submitted on : Wednesday, November 28, 2018 - 10:29:12 AM
Last modification on : Saturday, December 22, 2018 - 7:23:03 AM


Files produced by the author(s)


  • HAL Id : hal-01862201, version 1


Jyoti Maggu, Emilie Chouzenoux, Giovanni Chierchia, Angshul Majumdar. Convolutional Transform learning. 25th International Conference on Neural Information Processing (ICONIP 2018), Dec 2018, Siem Reap, Cambodia. pp.162-174. ⟨hal-01862201⟩



Record views


Files downloads