CMSN: Continuous Multi-stage Network and Variable Margin Cosine Loss for Temporal Action Proposal Generation
arXiv:1911.06080
Abstract
Accurately locating the start and end time of an action in untrimmed videos is a challenging task. One of the important reasons is the boundary of action is not highly distinguishable, and the features around the boundary are difficult to discriminate. To address this problem, we propose a novel framework for temporal action proposal generation, namely Continuous Multi-stage Network (CMSN), which divides a video that contains a complete action instance into six stages, namely Backgroud, Ready, Start, Confirm, End, Follow. To distinguish between Ready and Start, End and Follow more accurately, we propose a novel loss function, Variable Margin Cosine Loss (VMCL), which allows for different margins between different categories. Our experiments on THUMOS14 show that the proposed method for temporal proposal generation performs better than the state-of-the-art methods using the same network architecture and training dataset.
References in corpus (8)
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- Two-Stream Convolutional Networks for Action Recognition in Videos
- UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild
- Action Recognition with Trajectory-Pooled Deep-Convolutional Descriptors
- The THUMOS Challenge on Action Recognition for Videos "in the Wild"
- CUHK & ETHZ & SIAT Submission to ActivityNet Challenge 2016
- A Pursuit of Temporal Accuracy in General Activity Detection
- S3D: Single Shot multi-Span Detector via Fully 3D Convolutional Networks