Visual object tracking via LDA
Chih-Ching Lin, Shwu-Huey Yen, Ching‐Ting Tu · 2017
Tracking-by-detection methods treat the target location as a classification problem in which the approach SVM + HOG shows a good performance. However, training a good SVM classifier is cost expensive. In this paper, we replace SVM by linear discriminant analysis (LDA) for classification where the mean and covariance of negative examples are evaluated only once. Not only the training is much cheaper, but testing time is also very efficient. The proposed method uses HOG and color features. To defense partial occlusion issue, part-based tracking strategy is adopted and the model is updated according to Peak to Sidelobe Ratio (PSR) of parts. We evaluate our approach on several challenging video sequences, both qualitative and quantitative experiments show that our approach is competitive to stat-of-the-art methods and it can process 40-50 frames per second.