Active Learning for Structured Prediction from Partially Labeled Data

Abstract

We propose a general purpose active learning algorithm for structured prediction, gathering labeled data for training a model that outputs a set of related labels for an image or video. Active learning starts with a limited initial training set, then iterates querying a user for labels on unlabeled data and retraining the model. We propose a novel algorithm for selecting data for labeling, choosing examples to maximize expected information gain based on belief propagation inference. This is a general purpose method and can be applied to a variety of tasks or models. As a specific example we demonstrate this framework for learning to recognize human actions and group activities in video sequences. Experiments show that our proposed algorithm outperforms previous active learning methods and can achieve accuracy comparable to fully supervised methods while utilizing significantly less labeled data.

References in corpus (7)

• Caffe: Convolutional Architecture for Fast Feature Embedding
• Detecting events and key actors in multi-person videos
• Deep Structured Models For Group Activity Recognition
• Hierarchical Deep Temporal Models for Group Activity Recognition
• Tracking The Untrackable: Learning To Track Multiple Cues with Long-Term Dependencies
• Structure Inference Machines: Recurrent Neural Networks for Analyzing Relations in Group Activity Recognition
• Discovering Human Interactions in Videos with Limited Data Labeling