Temporal Convolution Based Action Proposal: Submission to ActivityNet 2017

TL;DR

This paper introduces a novel temporal convolutional network for action proposals that significantly improves the quality of temporal action localization in videos, achieving state-of-the-art results on ActivityNet.

Contribution

The paper presents a new temporal convolutional model specifically designed for action proposal generation, enhancing localization accuracy over previous methods.

Findings

Achieved state-of-the-art performance on ActivityNet proposal task.

Improved temporal action localization accuracy.

Validated effectiveness of temporal convolutional networks for proposals.

Abstract

In this notebook paper, we describe our approach in the submission to the temporal action proposal (task 3) and temporal action localization (task 4) of ActivityNet Challenge hosted at CVPR 2017. Since the accuracy in action classification task is already very high (nearly 90% in ActivityNet dataset), we believe that the main bottleneck for temporal action localization is the quality of action proposals. Therefore, we mainly focus on the temporal action proposal task and propose a new proposal model based on temporal convolutional network. Our approach achieves the state-of-the-art performances on both temporal action proposal task and temporal action localization task.