"Volumetric Semantic Segmentation using Pyramid Context Features" J. Barron, P. Arbelaez, S. Keranen, M. Biggin, D. Knowles and J. Malik.
Firstly, we propose a context measuring mechanism which could explicitly assess roles of context features for different object recognition tasks.
Add this to context features like room-size plus reverberation and a variety of gestures come out, while dealing with this sort of composition.
They also created context features that capture user's recent posts, both in semantic and emotional content, and their interactions with other users in the dataset.
In non-conversational interfaces, context features based on user interface system events can improve detection of head gestures for dialog box confirmation or document browsing.
The deep learning techniques can automatically learn texture and image context features from training data without the need of explicit feature engineering.
We propose a two-layer feature description structure that exploits the representation of spatio-temporal motion features and context features hierarchically.
This method takes the spatial relationship and temporal order of local features into account and creates the mid-level motion features and mid-level context features.
Using a discriminative approach to contextual prediction and multi-modal integration, performance of head gesture detection was improved with context features even when the topic of the test set was significantly different than the training set.
Since a clinical event is a basic unit of the problem and action relation, events are extracted from narrative texts, based on the external knowledge resources context features of the conditional random fields.
In the training stage, these extracted features are adopted to train a structured random forest classifier, which is further iteratively refined in an auto-context model by adopting the context features and the updated relationship features.
