Paper Info
Title
Generic performance metrics for continuous activity recognition
Abstract
For evaluating activity recognition results still classical error metrics like Accuracy, Precision, and Recall are being used. They are well understood and widely accepted but entail fundamental problems: They can not handle fuzzy event boundaries, or parallel activities, and they over-emphasize decision boundaries. We introduce more generic performance metrics as replacement, allowing for soft classification and annotation while being backward compatible. We argue that they can increase the expressiveness and still allow more sophisticated methods like event and segment analysis.
Year
DOI
Venue
2011
10.1007/978-3-642-24455-1_13
KI
Keywords
Field
DocType
over-emphasize decision boundary,continuous activity recognition,soft classification,generic performance metrics,segment analysis,fuzzy event boundary,fundamental problem,sophisticated method,parallel activity,activity recognition result,classical error metrics
Activity recognition,Annotation,Computer science,Fuzzy logic,Artificial intelligence,Recall,Machine learning,Backward compatibility,Expressivity
Conference
Volume
ISSN
Citations 
7006
0302-9743
2
PageRank 
References 
Authors
0.40
2
2
Name
Order
Citations
PageRank
Albert Hein1316.51
Thomas Kirste29318.37