EFFICIENT DECISION TREE TRAVERSALS IN AN ADAPTIVE BOOSTING (ADABOOST) CLASSIFIER

摘要

In described examples of a method for object classification in a decision tree based adaptive boosting (AdaBoost) classifier implemented on a single-instruction multiple-data (SIMD) processor, the method includes receiving (700) feature vectors extracted from N consecutive window positions in an image in a memory coupled to the SIMD processor and evaluating (708) the N consecutive window positions concurrently by the AdaBoost classifier using the feature vectors and vector instructions of the SIMD processor, in which the AdaBoost classifier concurrently traverses (714) decision trees for the N consecutive window positions until classification is complete (712) for the N consecutive window positions.