2020 29th International Conference on Computer Communications and Networks (ICCCN) 2020
DOI: 10.1109/icccn49398.2020.9209632
|View full text |Cite
|
Sign up to set email alerts
|

Minor Privacy Protection Through Real-time Video Processing at the Edge

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
5
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
6
1
1

Relationship

3
5

Authors

Journals

citations
Cited by 15 publications
(5 citation statements)
references
References 34 publications
0
5
0
Order By: Relevance
“…NICU face models demonstrate fine-tuning can enhance the performance of the robustness of new face detectors in complex NICU environments [33]. Umamageswari et al [34] comprehensively overviews some of the most effective DL-based face detection methods, categorizing them into a few major categories [35]. Addressing these challenges requires ongoing research and development in face identification.…”
Section: Literature Reviewmentioning
confidence: 99%
“…NICU face models demonstrate fine-tuning can enhance the performance of the robustness of new face detectors in complex NICU environments [33]. Umamageswari et al [34] comprehensively overviews some of the most effective DL-based face detection methods, categorizing them into a few major categories [35]. Addressing these challenges requires ongoing research and development in face identification.…”
Section: Literature Reviewmentioning
confidence: 99%
“…Unlike the DyCIE methods, the lightweight Regions of Interest Masking (RoI-Mask) scheme was developed based on the Peter De Jong Map [14] specifically for minor's privacy protection [39]. The Peter De Jong map is a type of 2D return f c recursive system stated in Eq.…”
Section: Regions Of Interest Maskingmentioning
confidence: 99%
“…However, for the sake of testing the security of the proposed mechanisms, the attackers in this testing model are assumed to have an access to a plain frame and its corresponding cipher so that they can perform visual assessment, differential analysis, histogram/frequency analysis, entropy analysis, and correlation analysis on the cipher and its corresponding clear form to find clues or weaknesses in the proposed algorithms. More details on the design rationales, function blocks, and evaluation reports of the edge VSS is beyond the scope of this study, interested readers can find details in literature [15,[38][39][40][41][42][43][44]. This study is about mechanisms designed to ensure E2E privacy in VSS, not about the surveillance per se.…”
Section: An Attacker Modelmentioning
confidence: 99%
“…In other words, video frame encryption is of paramount importance in ensuring E2E privacy in VSS. More details on the design rationales, function blocks, and evaluation reports of the edge VSS is beyond the scope of this study, interested readers can find details in literature[15,[38][39][40][41][42][43][44]. This study is about mechanisms designed to ensure E2E privacy in VSS, not about the surveillance per se.Privacy-protection schemes must be secure in that the reversion is done only by authorised parties who possess the right set of keys.…”
mentioning
confidence: 99%