Active learning based pedestrian detection in real scenes

This work presents an active learning based method for pedestrian detection in complicated real-world scenes. Through analyzing the distribution of all positive and negative samples under every possible feature, a highly efficient weak classifier selection method is presented. Moreover, a novel boosting architecture is given to get satisfied False Positive Rate (FPR) and False Negative Rate (FNR) with few weak classifiers. A unique characteristic of the algorithm is its ability to train special cascade classifier dynamically for each individual scene. The benefit is that the trained classifier will only focus on the differences between the positive samples and the limited negative samples of each individual scene, thus greatly reduce the complexity of classification and achieve robust detection result even with few classifiers. A real-time pedestrian detection system is developed based on the proposed algorithm. The system produces fast and robust detection results as demonstrated by extensive experiments which use video sequences under different environments.