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

We study the problem of future step anticipation in procedural videos. Givena video of an ongoing procedural activity, we predict a plausible nextprocedure step described in rich natural language. While most previous workfocus on the problem of data scarcity in procedural video datasets, anothercore challenge of future anticipation is how to account for multiple plausiblefuture realizations in natural settings. This problem has been largelyoverlooked in previous work. To address this challenge, we frame future stepprediction as modelling the distribution of all possible candidates for thenext step. Specifically, we design a generative model that takes a series ofvideo clips as input, and generates multiple plausible and diverse candidates(in natural language) for the next step. Following previous work, we side-stepthe video annotation scarcity by pretraining our model on a large text-basedcorpus of procedural activities, and then transfer the model to the videodomain. Our experiments, both in textual and video domains, show that our modelcaptures diversity in the next step prediction and generates multiple plausiblefuture predictions. Moreover, our model establishes new state-of-the-artresults on YouCookII, where it outperforms existing baselines on the next stepanticipation. Finally, we also show that our model can successfully transferfrom text to the video domain zero-shot, ie, without fine-tuning or adaptation,and produces good-quality future step predictions from video.

GePSAn: Generative Procedure Step Anticipation in Cooking Videos