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

Recent works have shown that the computational efficiency of videorecognition can be significantly improved by reducing the spatial redundancy.As a representative work, the adaptive focus method (AdaFocus) has achieved afavorable trade-off between accuracy and inference speed by dynamicallyidentifying and attending to the informative regions in each video frame.However, AdaFocus requires a complicated three-stage training pipeline(involving reinforcement learning), leading to slow convergence and isunfriendly to practitioners. This work reformulates the training of AdaFocus asa simple one-stage algorithm by introducing a differentiableinterpolation-based patch selection operation, enabling efficient end-to-endoptimization. We further present an improved training scheme to address theissues introduced by the one-stage formulation, including the lack ofsupervision, input diversity and training stability. Moreover, aconditional-exit technique is proposed to perform temporal adaptive computationon top of AdaFocus without additional training. Extensive experiments on sixbenchmark datasets (i.e., ActivityNet, FCVID, Mini-Kinetics,Something-Something V1&V2, and Jester) demonstrate that our model significantlyoutperforms the original AdaFocus and other competitive baselines, while beingconsiderably more simple and efficient to train. Code is available athttps://github.com/LeapLabTHU/AdaFocusV2.

AdaFocus V2: End-to-End Training of Spatial Dynamic Networks for Video Recognition