Weakly supervised vehicle detection in satellite images via multi-instance discriminative learning. (April 2017)

• Record Type: Journal Article
• Authors: Cao, Liujuan; Luo, Feng; Chen, Li; Sheng, Yihan; Wang, Haibin; Wang, Cheng; Ji, Rongrong
• Is Part Of: Pattern recognition. Volume 64(2017:Apr.)
• Page Start: 417
• Page End: 424
• Publication Date: 2017-04
• Subjects: Multiple instance learning -- Density estimation -- Multiple instance SVM -- Vehicle detection; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2016.10.033
• Languages: English
• ISSNs: 0031-3203

Abstract

Abstract: Vehicle detection in satellite images has attracted extensive research interest with widespreading application potentials. The main challenge lies in the difficulty of labeling sufficient training instances (vehicle rectangles) across all resolutions and imaging conditions of satellite images, which degenerates the performance of vehicle detectors trained correspondingly. To tackle this challenge, in this paper we propose an intelligent and labor-light scheme for large-scale training of vehicle detectors. Our scheme only requires region-level group annotation, i.e. whether this region contains vehicle(s) or not, without explicitly labeling the bounding boxes of vehicles. To this end, a novel weakly supervised, multi-instance learning algorithm is designed to learn instance-wise vehicle detectors from such "weak labels". In particular, a density estimator is firstly adopted to estimate the density map of vehicle instances from the positive regions. Then, a multi-instance SVM is trained to classify and locate vehicle instances from this map. We have carried out extensive experiments on a large-scale satellite image collection that contains various resolutions and imaging conditions. We have demonstrated that the proposed scheme has achieved superior performance by comparing to a set of state-of-the-art and alternative approaches. Abstract : Highlights: A user-friendly, scheme is proposed to efficiently collect large-scale vehicle annotations. A density estimation algorithm with integer programming is proposed to learn from weak labels to estimate the initial vehicle location. A large-margin classification termed MIL-SVM is adopted to learn refined vehicle detector.