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

Over the past few years, various tasks involving videos such asclassification, description, summarization and question answering have receiveda lot of attention. Current models for these tasks compute an encoding of thevideo by treating it as a sequence of images and going over every image in thesequence. However, for longer videos this is very time consuming. In thispaper, we focus on the task of video classification and aim to reduce thecomputational time by using the idea of distillation. Specifically, we firsttrain a teacher network which looks at all the frames in a video and computes arepresentation for the video. We then train a student network whose objectiveis to process only a small fraction of the frames in the video and stillproduce a representation which is very close to the representation computed bythe teacher network. This smaller student network involving fewer computationscan then be employed at inference time for video classification. We experimentwith the YouTube-8M dataset and show that the proposed student network canreduce the inference time by upto 30% with a very small drop in the performance

I Have Seen Enough: A Teacher Student Network for Video Classification Using Fewer Frames