Image-Level Automatic Data Augmentation for Pedestrian Detection

Data augmentation (DA) is a commonly used method to alleviate the problem of detecting occluded pedestrians in crowded scenes. Recently, several dataset-level automatic DA methods have been proposed to search for a set of general DA policies for the entire dataset, which saves a lot of time compared to manually designing DA policies. However, due to the huge differences between each image in pedestrian detection datasets, existing dataset-level augmentation methods cannot automatically adjust DA policies according to the differences between them, which will lead to outlier data and degradation of model performance. Therefore, considering the differences between each image, we propose an image-level automatic DA method that aims to find an optimal DA policy for each image in the dataset according to their respective characteristics. Specifically, we first reformulate the image-level automatic DA method by constructing a search space based on categorical distribution, within which we specify the probability of operations being sampled according to their respective effectiveness so that useful operations can be effectively preserved and useless operations can be suppressed. Subsequently, we design an encoding method to recode the index of images and policies and use the encoded index to closely associate them to achieve a stable matching relationship between images and policies. Finally, a search framework with Bayesian optimization is developed for efficient policy mining. Comprehensive experiments on CrowdHuman and CityPersons datasets show that compared with the commonly used automatic DA method for pedestrian detection, AutoPedestrian, our method takes only 1/14 of the search time, but achieves better detection accuracy. Specifically, we achieve 10.2% MR-2 on the CityPersons reasonable subset and 36.8% MR-2 on the CrowdHuman dataset, outperforming state-of-the-art methods on the CrowdHuman dataset.