Incremental Training of a Detector Using Online Sparse Eigen-decomposition

TL;DR

This paper introduces an efficient online learning framework for object detection using sparse eigen-decomposition, enabling adaptive updates without retraining from scratch, and demonstrates its effectiveness on handwriting and face datasets.

Contribution

The paper presents a novel online greedy sparse LDA framework for object detection, improving adaptability and efficiency over traditional offline and boosting methods.

Findings

Significant performance improvements in online object detection.

Robustness demonstrated on handwriting digit and face datasets.

Efficient online training reduces computational costs.

Abstract

The ability to efficiently and accurately detect objects plays a very crucial role for many computer vision tasks. Recently, offline object detectors have shown a tremendous success. However, one major drawback of offline techniques is that a complete set of training data has to be collected beforehand. In addition, once learned, an offline detector can not make use of newly arriving data. To alleviate these drawbacks, online learning has been adopted with the following objectives: (1) the technique should be computationally and storage efficient; (2) the updated classifier must maintain its high classification accuracy. In this paper, we propose an effective and efficient framework for learning an adaptive online greedy sparse linear discriminant analysis (GSLDA) model. Unlike many existing online boosting detectors, which usually apply exponential or logistic loss, our online algorithm makes use of LDA's learning criterion that not only aims to maximize the class-separation criterion but also incorporates the asymmetrical property of training data distributions. We provide a better alternative for online boosting algorithms in the context of training a visual object detector. We demonstrate the robustness and efficiency of our methods on handwriting digit and face data sets. Our results confirm that object detection tasks benefit significantly when trained in an online manner.