In this technical report, we describe our solution to temporal action proposal (task 1) in ActivityNet Challenge 2019. First, we fine-tune a ResNet-50-C3D CNN on ActivityNet v1.3 based on Kinetics pretrained model to extract snippet-level video representations and then we design a Relation-Aware Pyramid Network (RapNet) to generate temporal multiscale proposals with confidence score. After that, we employ a two-stage snippet-level boundary adjustment scheme to re-rank the order of generated proposals. Ensemble methods are also been used to improve the performance of our solution, which helps us achieve 2nd place.
Cite
@article{arxiv.1908.03448,
title = {Relation-Aware Pyramid Network (RapNet) for temporal action proposal},
author = {Jialin Gao and Zhixiang Shi and Jiani Li and Yufeng Yuan and Jiwei Li and Xi Zhou},
journal= {arXiv preprint arXiv:1908.03448},
year = {2019}
}
Comments
Submission to temporal action proposal task in ActivityNet Challenge 2019