Unsupervised Feature Learning from Temporal Data

Abstract

Current state-of-the-art classification and detection algorithms rely on supervised training. In this work we study unsupervised feature learning in the context of temporally coherent video data. We focus on feature learning from unlabeled video data, using the assumption that adjacent video frames contain semantically similar information. This assumption is exploited to train a convolutional pooling auto-encoder regularized by slowness and sparsity. We establish a connection between slow feature learning to metric learning and show that the trained encoder can be used to define a more temporally and semantically coherent metric.

arXiv admin note: substantial text overlap with arXiv:1412.6056

References in corpus (1)

• Saturating Auto-Encoders

Cited by in corpus (5)

• What makes ImageNet good for transfer learning?
• Soft + Hardwired Attention: An LSTM Framework for Human Trajectory Prediction and Abnormal Event Detection
• Object-Centric Representation Learning from Unlabeled Videos
• Learning Temporal Embeddings for Complex Video Analysis
• Ambient Sound Provides Supervision for Visual Learning