Project details for ABACOC Adaptive Ball Cover for Classification

Logo ABACOC Adaptive Ball Cover for Classification 1.0

by kikot - July 14, 2014, 16:27:03 CET [ BibTeX BibTeX for corresponding Paper Download ]

view (2 today), download ( 0 today ), 3 subscriptions

Description:

We introduce an online action recognition system that can be combined with any set of frame-by-frame feature descriptors. Our system covers the frame feature space with classifiers whose distribution adapts to the hardness of locally approximating the Bayes optimal classifier. An efficient nearest neighbour search is used to find and combine the local classifiers that are closest to the frames of a new video to be classified. The advantages of our approach are: incremental training, frame by frame real-time prediction, nonparametric predictive modelling, video segmentation for continuous action recognition, no need to trim videos to equal lengths and only one tuning parameter (which, for large datasets, can be safely set to the diameter of the feature space). Experiments on standard benchmarks show that our system is competitive with state-of-the-art non-incremental and incremental baselines.

Changes to previous version:

Initial release of the library, future changes will be advertised shortly.

BibTeX Entry: Download
Corresponding Paper BibTeX Entry: Download
Supported Operating Systems: Linux, Windows
Data Formats: Matlab, Java
Tags: Online Learning, Action Recognition, Incremental Learning, Multivariate Time Series Classification, Temporal Classification
Archive: download here

Comments

No one has posted any comments yet. Perhaps you'd like to be the first?

Leave a comment

You must be logged in to post comments.