Active Learning with Effective Scoring Functions for Semi-Supervised Temporal Action Localization. (arXiv:2208.14856v1 [cs.CV])

Temporal Action Localization (TAL) aims to predict both action category and
temporal boundary of action instances in untrimmed videos, i.e., start and end
time. Fully-supervised solutions are usually adopted in most existing works,
and proven to be effective. One of the practical bottlenecks in these solutions
is the large amount of labeled training data required. To reduce expensive
human label cost, this paper focuses on a rarely investigated yet practical
task named semi-supervised TAL and proposes an effective active learning …

active learning arxiv learning localization scoring semi-supervised temporal