Extensive adoption of video surveillance, affecting many aspects of the daily life, alarms the concerned public about the increasing invasion into personal privacy. To address these concerns, many tools have been proposed for protection of personal privacy in image and video. However, little is understood regarding the effectiveness of such tools and especially their impact on the underlying surveillance tasks. In this paper, we propose conducting a subjective evaluation using crowdsourcing to analyze the tradeoff between the preservation of privacy offered by these tools and the intelligibility of activities under video surveillance. As an example, the proposed method is used to compare several commonly employed privacy protection techniques, such as blurring, pixelization, and masking applied to indoor surveillance video. Facebook based crowdsourcing application was specifically developed to gather the subjective evaluation data. Based on more than one hundred participants, the evaluation results demonstrate that the pixelization filter provides the best performance in terms of balance between privacy protection and intelligibility. The results obtained with crowdsourcing application were compared with results of previous work using more conventional subjective tests showing that they are highly correlated.
{"title":"Crowdsourcing approach for evaluation of privacy filters in video surveillance","authors":"Pavel Korshunov, Shuting Cai, T. Ebrahimi","doi":"10.1145/2390803.2390817","DOIUrl":"https://doi.org/10.1145/2390803.2390817","url":null,"abstract":"Extensive adoption of video surveillance, affecting many aspects of the daily life, alarms the concerned public about the increasing invasion into personal privacy. To address these concerns, many tools have been proposed for protection of personal privacy in image and video. However, little is understood regarding the effectiveness of such tools and especially their impact on the underlying surveillance tasks. In this paper, we propose conducting a subjective evaluation using crowdsourcing to analyze the tradeoff between the preservation of privacy offered by these tools and the intelligibility of activities under video surveillance. As an example, the proposed method is used to compare several commonly employed privacy protection techniques, such as blurring, pixelization, and masking applied to indoor surveillance video. Facebook based crowdsourcing application was specifically developed to gather the subjective evaluation data. Based on more than one hundred participants, the evaluation results demonstrate that the pixelization filter provides the best performance in terms of balance between privacy protection and intelligibility. The results obtained with crowdsourcing application were compared with results of previous work using more conventional subjective tests showing that they are highly correlated.","PeriodicalId":429491,"journal":{"name":"CrowdMM '12","volume":"334-335 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130915378","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Crowdsourcing is becoming increasingly popular as a cheap and effective tool for multimedia annotation. However, the idea is not new, and can be traced back to Charles Darwin. He was interested in studying the universality of facial expressions in conveying emotions, thus he had to consider a global population. Access to different cultures allowed him to reach more general conclusions. In this paper, we highlight a few milestones in the history of the study of emotion that share the concepts of crowdsourcing. We first consider the study of posed photographs and then move to videos of natural expressions. We present our use of crowdsouring to label a video corpus of natural expressions, and also to recreate one of Darwin's original emotion judgment experiments. This allows us to compare people's perception of emotional expressions in the 19th and 21st centuries, showing that it remains stable through both culture and time.
{"title":"Crowdsouring in emotion studies across time and culture","authors":"M. Mahmoud, T. Baltrušaitis, P. Robinson","doi":"10.1145/2390803.2390810","DOIUrl":"https://doi.org/10.1145/2390803.2390810","url":null,"abstract":"Crowdsourcing is becoming increasingly popular as a cheap and effective tool for multimedia annotation. However, the idea is not new, and can be traced back to Charles Darwin. He was interested in studying the universality of facial expressions in conveying emotions, thus he had to consider a global population. Access to different cultures allowed him to reach more general conclusions. In this paper, we highlight a few milestones in the history of the study of emotion that share the concepts of crowdsourcing. We first consider the study of posed photographs and then move to videos of natural expressions. We present our use of crowdsouring to label a video corpus of natural expressions, and also to recreate one of Darwin's original emotion judgment experiments. This allows us to compare people's perception of emotional expressions in the 19th and 21st centuries, showing that it remains stable through both culture and time.","PeriodicalId":429491,"journal":{"name":"CrowdMM '12","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130758191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：