In this paper, we introduce our submissions for the tasks of trimmed activity recognition (Kinetics) and trimmed event recognition (Moments in Time) for Activitynet Challenge 2018. In the two tasks, non-local neural networks and temporal segment networks are implemented as our base models. Multi-modal cues such as RGB image, optical flow and acoustic signal have also been used in our method. We also propose new non-local-based models for further improvement on the recognition accuracy. The final submissions after ensembling the models achieve 83.5% top-1 accuracy and 96.8% top-5 accuracy on the Kinetics validation set, 35.81% top-1 accuracy and 62.59% top-5 accuracy on the MIT validation set.
@article{arxiv.1806.04391,
title = {Qiniu Submission to ActivityNet Challenge 2018},
author = {Xiaoteng Zhang and Yixin Bao and Feiyun Zhang and Kai Hu and Yicheng Wang and Liang Zhu and Qinzhu He and Yining Lin and Jie Shao and Yao Peng},
journal= {arXiv preprint arXiv:1806.04391},
year = {2018}
}