English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep neural networks need a big amount of training data, while in the realworld there is a scarcity of data available for training purposes. To resolvethis issue unsupervised methods are used for training with limited data. Inthis report, we describe the unsupervised semantic feature learning approachfor recognition of the geometric transformation applied to the input data. Thebasic concept of our approach is that if someone is unaware of the objects inthe images, he/she would not be able to quantitatively predict the geometrictransformation that was applied to them. This self supervised scheme is basedon pretext task and the downstream task. The pretext classification task toquantify the geometric transformations should force the CNN to learn high-levelsalient features of objects useful for image classification. In our baselinemodel, we define image rotations by multiples of 90 degrees. The CNN trained onthis pretext task will be used for the classification of images in the CIFAR-10dataset as a downstream task. we run the baseline method using various models,including ResNet, DenseNet, VGG-16, and NIN with a varied number of rotationsin feature extracting and fine-tuning settings. In extension of this baselinemodel we experiment with transformations other than rotation in pretext task.We compare performance of selected models in various settings with differenttransformations applied to images,various data augmentation techniques as wellas using different optimizers. This series of different type of experimentswill help us demonstrate the recognition accuracy of our self-supervised modelwhen applied to a downstream task of classification.

TransformNet: Self-supervised representation learning through predicting geometric transformations