Systems and methods for object classification, object detection and memory management

摘要

A method for object classification by an electronic device is described. The method includes obtaining an image frame that includes an object. The method also includes determining samples from the image frame. Each of the samples represents a multidimensional feature vector. The method further includes adding the samples to a training set for the image frame. The method additionally includes pruning one or more samples from the training set to produce a pruned training set. One or more non-support vector negative samples are pruned first. One or more non-support vector positive samples are pruned second if necessary to avoid exceeding a sample number threshold. One or more support vector samples are pruned third if necessary to avoid exceeding the sample number threshold. The method also includes updating classifier model weights based on the pruned training set.

主权项

1. A method for object classification by an electronic device, comprising:
obtaining an image frame that includes an object; determining samples from the image frame, wherein each of the samples represents a multidimensional feature vector, wherein the samples comprise non-support vector negative samples, non-support vector positive samples, and support vector samples; adding the samples to a training set for the image frame; pruning one or more samples from the training set to produce a pruned training set, wherein one or more of the non-support vector negative samples are pruned first, wherein one or more of the non-support vector positive samples are pruned second based on a comparison between a number of samples remaining after pruning the one or more non-support vector negative samples and a sample number threshold, and wherein one or more of the support vector samples are pruned third if necessary to avoid exceeding the sample number threshold; and updating classifier model weights based on the pruned training set.