{"title":"基于时空加权多实例学习的鲁棒跟踪","authors":"Li Wang, Xiao'an Tang, Dongdong Li","doi":"10.1109/DICTA.2017.8227488","DOIUrl":null,"url":null,"abstract":"Due to the superiority in handling label ambiguity, multiple instance learning (MIL) has been introduced into adaptive tracking-by-detection methods to alleviate drift and yields promising tracking performance. However, the MIL tracker assumes that all samples in a positive bag contribute equally to the bag probability, which ignores sample importance. To address this issue, in this paper we propose a spatio- temporally weighted MIL (STWMIL) tracker which integrates temporal weight into the update scheme for Haar-like features and spatial weight into the bag probability function. Spatial weight for the positive sample near the target location is larger than that far from the target location, which means the former contributes more to the positive bag probability. Based on spatial weight, a novel bag probability function is proposed using the weighted Noisy-OR model. Temporal weight for the recently-acquired images is larger than that for the earlier observations, which means less modeling power is expended on old observations. Based on temporal weight, a novel update scheme with changing but convergent learning rate is derived with strict mathematic proof. Extensive experiments performed on the OTB-2013 tracking benchmark demonstrate that our proposed tracker achieves superior performance both qualitatively and quantitatively over several state-of- the-art trackers.","PeriodicalId":194175,"journal":{"name":"2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Robust Tracking via Spatio-Temporally Weighted Multiple Instance Learning\",\"authors\":\"Li Wang, Xiao'an Tang, Dongdong Li\",\"doi\":\"10.1109/DICTA.2017.8227488\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Due to the superiority in handling label ambiguity, multiple instance learning (MIL) has been introduced into adaptive tracking-by-detection methods to alleviate drift and yields promising tracking performance. However, the MIL tracker assumes that all samples in a positive bag contribute equally to the bag probability, which ignores sample importance. To address this issue, in this paper we propose a spatio- temporally weighted MIL (STWMIL) tracker which integrates temporal weight into the update scheme for Haar-like features and spatial weight into the bag probability function. Spatial weight for the positive sample near the target location is larger than that far from the target location, which means the former contributes more to the positive bag probability. Based on spatial weight, a novel bag probability function is proposed using the weighted Noisy-OR model. Temporal weight for the recently-acquired images is larger than that for the earlier observations, which means less modeling power is expended on old observations. Based on temporal weight, a novel update scheme with changing but convergent learning rate is derived with strict mathematic proof. Extensive experiments performed on the OTB-2013 tracking benchmark demonstrate that our proposed tracker achieves superior performance both qualitatively and quantitatively over several state-of- the-art trackers.\",\"PeriodicalId\":194175,\"journal\":{\"name\":\"2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA)\",\"volume\":\"11 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/DICTA.2017.8227488\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/DICTA.2017.8227488","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Robust Tracking via Spatio-Temporally Weighted Multiple Instance Learning
Due to the superiority in handling label ambiguity, multiple instance learning (MIL) has been introduced into adaptive tracking-by-detection methods to alleviate drift and yields promising tracking performance. However, the MIL tracker assumes that all samples in a positive bag contribute equally to the bag probability, which ignores sample importance. To address this issue, in this paper we propose a spatio- temporally weighted MIL (STWMIL) tracker which integrates temporal weight into the update scheme for Haar-like features and spatial weight into the bag probability function. Spatial weight for the positive sample near the target location is larger than that far from the target location, which means the former contributes more to the positive bag probability. Based on spatial weight, a novel bag probability function is proposed using the weighted Noisy-OR model. Temporal weight for the recently-acquired images is larger than that for the earlier observations, which means less modeling power is expended on old observations. Based on temporal weight, a novel update scheme with changing but convergent learning rate is derived with strict mathematic proof. Extensive experiments performed on the OTB-2013 tracking benchmark demonstrate that our proposed tracker achieves superior performance both qualitatively and quantitatively over several state-of- the-art trackers.