Online Detection of Action Start in Untrimmed, Streaming Videos
arXiv:1802.06822
Abstract
We aim to tackle a novel task in action detection - Online Detection of Action Start (ODAS) in untrimmed, streaming videos. The goal of ODAS is to detect the start of an action instance, with high categorization accuracy and low detection latency. ODAS is important in many applications such as early alert generation to allow timely security or emergency response. We propose three novel methods to specifically address the challenges in training ODAS models: (1) hard negative samples generation based on Generative Adversarial Network (GAN) to distinguish ambiguous background, (2) explicitly modeling the temporal consistency between data around action start and data succeeding action start, and (3) adaptive sampling strategy to handle the scarcity of training data. We conduct extensive experiments using THUMOS'14 and ActivityNet. We show that our proposed methods lead to significant performance gains and improve the state-of-the-art methods. An ablation study confirms the effectiveness of each proposed method.
arXiv commentAccepted by ECCV'18
References in corpus (20)
- Adam: A Method for Stochastic Optimization
- TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- Two-Stream Convolutional Networks for Action Recognition in Videos
- Conditional Image Synthesis With Auxiliary Classifier GANs
- Going Deeper with Convolutions
- Convolutional Two-Stream Network Fusion for Video Action Recognition
- ConvNet Architecture Search for Spatiotemporal Feature Learning
- Good Semi-supervised Learning that Requires a Bad GAN
- Temporal Action Detection with Structured Segment Networks
- R-C3D: Region Convolutional 3D Network for Temporal Activity Detection
- Advances in Human Action Recognition: A Survey
- TALL: Temporal Activity Localization via Language Query
- Every Moment Counts: Dense Detailed Labeling of Actions in Complex Videos
- TURN TAP: Temporal Unit Regression Network for Temporal Action Proposals
- Cascaded Boundary Regression for Temporal Action Detection
- Review of Action Recognition and Detection Methods
- Predictive-Corrective Networks for Action Detection
- Encouraging LSTMs to Anticipate Actions Very Early
- Spatio-Temporal Action Detection with Cascade Proposal and Location Anticipation