Skip-Clip: Self-Supervised Spatiotemporal Representation Learning by Future Clip Order Ranking

Abstract

Deep neural networks require collecting and annotating large amounts of data to train successfully. In order to alleviate the annotation bottleneck, we propose a novel self-supervised representation learning approach for spatiotemporal features extracted from videos. We introduce Skip-Clip, a method that utilizes temporal coherence in videos, by training a deep model for future clip order ranking conditioned on a context clip as a surrogate objective for video future prediction. We show that features learned using our method are generalizable and transfer strongly to downstream tasks. For action recognition on the UCF101 dataset, we obtain 51.8% improvement over random initialization and outperform models initialized using inflated ImageNet parameters. Skip-Clip also achieves results competitive with state-of-the-art self-supervision methods.

Holistic Video Understanding Workshop ICCV2019

References in corpus (8)

• Adam: A Method for Stochastic Optimization
• Deep Residual Learning for Image Recognition
• UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild
• Generating Videos with Scene Dynamics
• Unsupervised Visual Representation Learning by Context Prediction
• Unsupervised Learning of Visual Representations using Videos
• Self-Supervised Spatiotemporal Feature Learning via Video Rotation Prediction
• Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?

Cited by in corpus (4)

• Self-Supervised Representation Learning for Detection of ACL Tear Injury in Knee MR Videos
• Self-Supervised Learning via multi-Transformation Classification for Action Recognition
• There Is No Turning Back: A Self-Supervised Approach for Reversibility-Aware Reinforcement Learning
• MIO : Mutual Information Optimization using Self-Supervised Binary Contrastive Learning