Progression-Guided Temporal Action Detection in Videos

TL;DR

This paper introduces the Action Progression Network (APN), a novel end-to-end framework for temporal action detection in videos that encodes action evolution into 101 stages to improve detection accuracy, especially for long-lasting actions.

Contribution

The paper proposes a new action progression encoding method and an end-to-end neural network framework for more accurate and robust temporal action detection in videos.

Findings

Achieves 58.3% mAP on THUMOS14 at IoU 0.5

Achieves 98.9% mAP on DFMAD70 dataset

Effectively detects long-lasting actions and avoids incomplete detections

Abstract

We present a novel framework, Action Progression Network (APN), for temporal action detection (TAD) in videos. The framework locates actions in videos by detecting the action evolution process. To encode the action evolution, we quantify a complete action process into 101 ordered stages (0\%, 1\%, ..., 100\%), referred to as action progressions. We then train a neural network to recognize the action progressions. The framework detects action boundaries by detecting complete action processes in the videos, e.g., a video segment with detected action progressions closely follow the sequence 0\%, 1\%, ..., 100\%. The framework offers three major advantages: (1) Our neural networks are trained end-to-end, contrasting conventional methods that optimize modules separately; (2) The APN is trained using action frames exclusively, enabling models to be trained on action classification datasets and robust to videos with temporal background styles differing from those in training; (3) Our framework effectively avoids detecting incomplete actions and excels in detecting long-lasting actions due to the fine-grained and explicit encoding of the temporal structure of actions. Leveraging these advantages, the APN achieves competitive performance and significantly surpasses its counterparts in detecting long-lasting actions. With an IoU threshold of 0.5, the APN achieves a mean Average Precision (mAP) of 58.3\% on the THUMOS14 dataset and 98.9\% mAP on the DFMAD70 dataset.