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

Self-supervised learning (SSL) methods such as masked language modeling haveshown massive performance gains by pretraining transformer models for a varietyof natural language processing tasks. The follow-up research adapted similarmethods like masked image modeling in vision transformer and demonstratedimprovements in the image classification task. Such simple self-supervisedmethods are not exhaustively studied for object detection transformers (DETR,Deformable DETR) as their transformer encoder modules take input in theconvolutional neural network (CNN) extracted feature space rather than theimage space as in general vision transformers. However, the CNN feature mapsstill maintain the spatial relationship and we utilize this property to designself-supervised learning approaches to train the encoder of object detectiontransformers in pretraining and multi-task learning settings. We explore commonself-supervised methods based on image reconstruction, masked image modelingand jigsaw. Preliminary experiments in the iSAID dataset demonstrate fasterconvergence of DETR in the initial epochs in both pretraining and multi-tasklearning settings; nonetheless, similar improvement is not observed in the caseof multi-task learning with Deformable DETR. The code for our experiments withDETR and Deformable DETR are available at https://github.com/gokulkarthik/detrand https://github.com/gokulkarthik/Deformable-DETR respectively.

An Empirical Study Of Self-supervised Learning Approaches For Object Detection With Transformers