Phylogenetic pattern provides one way to estimate and quantify variation at the feature level.
Conversely, we incorporate colour and depth cues at the feature level within a Bayesian framework.
Results indicate that combining MODGDF with MFCC at the feature level gives significant improvements for speech recognition tasks in noise.
It is not easy to combine the information at the feature level when the feature sets obtained by different techniques are either inaccessible or incompatible.
In this approach, an information fusion method is used to combine base digital image forensic models at the feature level and the decision level.
Furthermore, according to the preserved magnitude, a comparative competitive code (C2Code) is explored for finger vein and dorsal fusion at the feature level.
In this paper, we present a multimodal personal identification system that fuses finger vein and finger dorsal images at the feature level.
In general, fusion at the feature level is easy and straightforward to implement, with the simplest case being the concatenation of features across different modalities (i.e., audio, video) into a single vector.
Moreover, extracted elements of the scene can be grouped into higher level ones, such as crossroads, bridges and overpasses, through data fusion at the feature level, because the procedure is characterized by a multi-scale, object-based approach.
This process could occur not only at the feature level, which facilitates target identification via feature updating (the updated feature is identical to the target in the feature), but also at the response level, which benefits target response via response link updating (an updated response link has an identical reaction to the target).
