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

Temporal Action Localization (TAL) methods typically operate on top offeature sequences from a frozen snippet encoder that is pretrained with theTrimmed Action Classification (TAC) tasks, resulting in a task discrepancyproblem. While existing TAL methods mitigate this issue either by retrainingthe encoder with a pretext task or by end-to-end fine-tuning, they commonlyrequire an overload of high memory and computation. In this work, we introduceSoft-Landing (SoLa) strategy, an efficient yet effective framework to bridgethe transferability gap between the pretrained encoder and the downstream tasksby incorporating a light-weight neural network, i.e., a SoLa module, on top ofthe frozen encoder. We also propose an unsupervised training scheme for theSoLa module; it learns with inter-frame Similarity Matching that uses the frameinterval as its supervisory signal, eliminating the need for temporalannotations. Experimental evaluation on various benchmarks for downstream TALtasks shows that our method effectively alleviates the task discrepancy problemwith remarkable computational efficiency.

Soft-Landing Strategy for Alleviating the Task Discrepancy Problem in Temporal Action Localization Tasks