EU Court of Justice determines that data anonymisation “might work, but then again it might not”

It follows from the provisions of that regulation as interpreted  in case-law that pseudonymisation may, depending on the circumstances of the case, effectively prevent persons other than the controller from identifying the data subject in such a way that, for them, the data subject is not or is no longer identifiable. In that context, the Court of Justice is careful to recall the guidance…

from its case-law regarding the assessment of whether or not the data subject is identifiable in situations in which the information enabling that subject to be identified was not in the hands of other people.

https://curia.europa.eu/jcms/upload/docs/application/pdf/2025-09/cp250107en.pdf

#anonymisation #dataProtection #eu #linkedData #pii #privacy #reidentification

In recent years, the development and refinement of #artificialintelligence (AI) technologies have revolutionised #computervision. This #TemaEU article explores the usage of #syntheticimages in training and improving #peopledetection and #reidentification models, highlighting its benefits, challenges, and future implications.

👉 https://tema-project.eu/articles/synthetic-images-people-detection-balancing-effectiveness-and-ethical-imperatives

Our latest work "Neural Texture Puppeteer" is published at https://openaccess.thecvf.com/content/WACV2024W/CV4Smalls/html/Waldmann_Neural_Texture_Puppeteer_A_Framework_for_Neural_Geometry_and_Texture_WACVW_2024_paper.html

As a base we make use of "Neural Puppeteer", an efficient and flexible neural rendering pipeline https://openaccess.thecvf.com/content/ACCV2022/html/Giebenhain_Neural_Puppeteer_Keypoint-Based_Neural_Rendering_of_Dynamic_Shapes_ACCV_2022_paper.html

Our key idea is to disentangle texture and geometry.

We show with twelve distinct synthetic cow textures that the new pipeline can be used in a downstream task to identify individuals.

#NeTePu #NePu #WACV #WACV24 #computervision @unikonstanz #CBehav #NeuralRendering #ReIdentification

"Correct identification rates were typically 86–100%, indicating a high risk of reidentification. "

Does deidentification of data from wearable devices give us a false sense of security? A systematic review

https://www.sciencedirect.com/science/article/pii/S2589750022002345

(article is Open)
#Reidentification
#WearableMedicalDevices

I am truly curious if I'm capable of getting through a data-related presentation without invoking Latanya Sweeney's research.

#KAnonymity
#Reidentification
#DeidentifiedMyLeftEarlobe
#Data

"Estimating the success of #reidentification in incomplete datasets":

"We find that 99.98% of Americans would be correctly re-identified in any dataset using 15 demographic attributes. Our results suggest that even heavily sampled anonymized #datasets are unlikely to satisfy the modern standards for #anonymization set forth by #GDPR and seriously challenge the technical and legal adequacy of the de-identification release-and-forget model."
⸺
https://www.nature.com/articles/s41467-019-10933-3/ @dataGovernance #dataPrivacy

Does #deidentification of #data from #wearable devices give us a false sense of #security? A #SystematicReview

#OpenAccess

"This systematic review seeks to investigate whether deidentifying data from wearable devices is sufficient to protect the privacy of individuals in datasets. Correct identification rates were typically 86–100%, indicating a high risk of #reidentification. As little as 1–300 s of recording were required to enable reidentification."

https://www.thelancet.com/journals/landig/article/PIIS2589-7500(22)00234-5/fulltext