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

Recognizing the activities causing distraction in real-world drivingscenarios is critical for ensuring the safety and reliability of both driversand pedestrians on the roadways. Conventional computer vision techniques aretypically data-intensive and require a large volume of annotated training datato detect and classify various distracted driving behaviors, thereby limitingtheir efficiency and scalability. We aim to develop a generalized frameworkthat showcases robust performance with access to limited or no annotatedtraining data. Recently, vision-language models have offered large-scalevisual-textual pretraining that can be adapted to task-specific learning likedistracted driving activity recognition. Vision-language pretraining models,such as CLIP, have shown significant promise in learning naturallanguage-guided visual representations. This paper proposes a CLIP-based driveractivity recognition approach that identifies driver distraction fromnaturalistic driving images and videos. CLIP's vision embedding offerszero-shot transfer and task-based finetuning, which can classify distractedactivities from driving video data. Our results show that this framework offersstate-of-the-art performance on zero-shot transfer and video-based CLIP forpredicting the driver's state on two public datasets. We propose bothframe-based and video-based frameworks developed on top of the CLIP's visualrepresentation for distracted driving detection and classification tasks andreport the results.

Vision-Language Models can Identify Distracted Driver Behavior from Naturalistic Videos