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

Traditional temporal action detection (TAD) usually handles untrimmed videoswith small number of action instances from a single label (e.g., ActivityNet,THUMOS). However, this setting might be unrealistic as different classes ofactions often co-occur in practice. In this paper, we focus on the task ofmulti-label temporal action detection that aims to localize all actioninstances from a multi-label untrimmed video. Multi-label TAD is morechallenging as it requires for fine-grained class discrimination within asingle video and precise localization of the co-occurring instances. Tomitigate this issue, we extend the sparse query-based detection paradigm fromthe traditional TAD and propose the multi-label TAD framework of PointTAD.Specifically, our PointTAD introduces a small set of learnable query points torepresent the important frames of each action instance. This point-basedrepresentation provides a flexible mechanism to localize the discriminativeframes at boundaries and as well the important frames inside the action.Moreover, we perform the action decoding process with the Multi-levelInteractive Module to capture both point-level and instance-level actionsemantics. Finally, our PointTAD employs an end-to-end trainable frameworksimply based on RGB input for easy deployment. We evaluate our proposed methodon two popular benchmarks and introduce the new metric of detection-mAP formulti-label TAD. Our model outperforms all previous methods by a large marginunder the detection-mAP metric, and also achieves promising results under thesegmentation-mAP metric. Code is available athttps://github.com/MCG-NJU/PointTAD.

PointTAD: Multi-Label Temporal Action Detection with Learnable Query Points