Selective Weakly Supervised Human Detection under Arbitrary Poses. (May 2017)

• Record Type: Journal Article
• Authors: Cai, Yawei; Tan, Xiaosong; Tan, Xiaoyang
• Is Part Of: Pattern recognition. Volume 65(2017:May)
• Page Start: 223
• Page End: 237
• Publication Date: 2017-05
• Subjects: Weakly supervised learning -- Human detection -- Selective Weakly Supervised Detection (SWSD) -- Multi-instance learning (MIL); Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2016.12.025
• Languages: English
• ISSNs: 0031-3203

Abstract

Abstract: In this paper we study the problem of weakly supervised human detection under arbitrary poses within the framework of multi-instance learning (MIL). Our contributions are threefold: (1) we first show that in the context of weakly supervised learning, some commonly used bagging tools in MIL such as the Noisy-OR model or the ISR model tend to suffer from the problem of gradient magnitude reduction when the initial instance-level detector is weak and/or when there exist large number of negative proposals, resulting in extremely inefficient use of training examples. We hence advocate the use of more robust and simple max-pooling rule or average rule under such circumstances; (2) we propose a new Selective Weakly Supervised Detection (SWSD) algorithm, which is shown to outperform several previous state-of-the-art weakly supervised methods; (3) finally, we identify several crucial factors that may significantly influence the performance, such as the usefulness of a small amount of supervision information, the need of relatively higher RoP (Ratio of Positive Instances), and so on – these factors are shown to benefit the MIL-based weakly supervised detector but are less studied in the previous literature. We also annotate a new large-scale data set called LSP/MPII-MPHB (Multiple Poses Human Body), in which and another popular benchmark dataset we demonstrate the superiority of the proposed method compared to several previous state-of-the-art methods. Abstract : Highlights: We propose a novel Selective Weakly Supervised Detection method which outperforms the previous state-of-the-art methods. We annotate a new large-scale data set called LSP/MPII-MPHB (Multiple Poses Human Body) for human body detection. We identify an easily ignored pitfall of the Noisy-OR model in MIL, which can significantly reduce the training efficiency of a MIL algorithm. We present a comprehensive and in-depth empirical study of the weakly supervised MIL method.