paper-with-me

Papers

Does Image Anonymization Impact Computer Vision Training?

2023-06-08 · Håkon Hukkelås, Frank Lindseth

Image anonymization is widely adapted in practice to comply with privacy regulations in many regions. However, anonymization often degrades the quality of the data, reducing its utility for computer vision development. In this paper, we investigate the impact of image anonymization for training computer vision models on key computer vision tasks (detection, instance segmentation, and pose estimation). Specifically, we benchmark the recognition drop on common detection datasets, where we evaluate both traditional and realistic anonymization for faces and full bodies. Our comprehensive experiments reflect that traditional image anonymization substantially impacts final model performance, particularly when anonymizing the full body. Furthermore, we find that realistic anonymization can mitigate this decrease in performance, where our experiments reflect a minimal performance drop for face anonymization. Our study demonstrates that realistic anonymization can enable privacy-preserving computer vision development with minimal performance degradation across a range of important computer vision benchmarks.

📄 PDF Abstract BibTeX arXiv:2306.05135

Code (1)

hukkelas/deep_privacy2 공식 구현 pytorch

Tasks

Face AnonymizationInstance SegmentationPose EstimationPrivacy PreservingSemantic Segmentation

Similar Papers 제목 키워드 기반

Evaluating the Impact of Data Anonymization on Image Retrieval

2026-02-23 · Marvin Chen, Manuel Eberhardinger, Johannes Maucher arxiv

With the growing importance of privacy regulations such as the General Data Protection Regulation, anonymizing visual data is becoming increasingly relevant across institutions. However, anonymization can negatively affe…

Content-Based Image Retrieval

CIAGAN: Conditional Identity Anonymization Generative Adversarial Networks

2020-05-19 · CVPR 2020 6 · Maxim Maximov, Ismail Elezi, Laura Leal-Taixé

The unprecedented increase in the usage of computer vision technology in society goes hand in hand with an increased concern in data privacy. In many real-world scenarios like people tracking or action recognition, it is…

Action RecognitionDe-identificationDiversityFace Anonymization

Controllable Localized Face Anonymization Via Diffusion Inpainting

2025-09-18 · Ali Salar, Qing Liu, Guoying Zhao arxiv

The growing use of portrait images in computer vision highlights the need to protect personal identities. At the same time, anonymized images must remain useful for downstream computer vision tasks. In this work, we prop…

Face Anonymization

Realistic Full-Body Anonymization with Surface-Guided GANs

2022-01-06 · Håkon Hukkelås, Morten Smebye, Rudolf Mester, Frank Lindseth

Recent work on image anonymization has shown that generative adversarial networks (GANs) can generate near-photorealistic faces to anonymize individuals. However, scaling up these networks to the entire human body has re…

DiversityFull-body anonymization

Password-conditioned Anonymization and Deanonymization with Face Identity Transformers

2019-11-26 · Xiuye Gu, Weixin Luo, Michael S. Ryoo, Yong Jae Lee

Cameras are prevalent in our daily lives, and enable many useful systems built upon computer vision technologies such as smart cameras and home robots for service applications. However, there is also an increasing societ…