mlsanityData to 7 Oct 2026

A Study of Face Obfuscation in ImageNet

arXiv 2103.06191 ↗First posted 10 Mar 2021cs.CVcited by 83

Not on the map. Why?

Abstract (arXiv)

Face obfuscation (blurring, mosaicing, etc.) has been shown to be effective for privacy protection; nevertheless, object recognition research typically assumes access to complete, unobfuscated images. In this paper, we explore the effects of face obfuscation on the popular ImageNet challenge visual recognition benchmark. Most categories in the ImageNet challenge are not people categories; however, many incidental people appear in the images, and their privacy is a concern. We first annotate faces in the dataset. Then we demonstrate that face obfuscation has minimal impact on the accuracy of recognition models. Concretely, we benchmark multiple deep neural networks on obfuscated images and observe that the overall recognition accuracy drops only slightly (<= 1.0%). Further, we experiment with transfer learning to 4 downstream tasks (object recognition, scene recognition, face attribute classification, and object detection) and show that features learned on obfuscated images are equally transferable. Our work demonstrates the feasibility of privacy-aware visual recognition, improves the highly-used ImageNet challenge benchmark, and suggests an important path for future visual datasets. Data and code are available at https://github.com/princetonvisualai/imagenet-face-obfuscation.

Key facts. Cited by 83 papers in the map (data through 7 Oct 2026). Builds on 59 of its 100 references that are in the map. First posted on arXiv 10 Mar 2021.

Builds on

59 of 100 references.

References · 59

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  5. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  6. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  7. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  8. 2016Densely Connected Convolutional Networkscited by 6,691
  9. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  10. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  11. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  12. 2014Going Deeper with Convolutionscited by 4,231
38 more
  1. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
  2. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  3. 2016Wide Residual Networkscited by 3,057
  4. 2015Fast R-CNNcited by 2,930
  5. 2016Deep Learning with Differential Privacycited by 2,682
  6. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,540
  7. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
  8. 2020Extracting Training Data from Large Language Modelscited by 2,144
  9. 2019Federated Learning: Challenges, Methods, and Future Directionscited by 1,531
  10. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,379
  11. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,350
  12. 2017ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devicescited by 1,194
  13. 2016SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model sizecited by 1,054
  14. 2018MnasNet: Platform-Aware Neural Architecture Search for Mobilecited by 937
  15. 2018The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networkscited by 847
  16. 2018Exploiting Unintended Feature Leakage in Collaborative Learningcited by 509
  17. 2017Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraintscited by 420
  18. 2020Data and its (dis)contents: A survey of dataset development and use in machine learning researchcited by 259
  19. 2020Large image datasets: A pyrrhic win for computer vision?cited by 237
  20. 2018Women also Snowboard: Overcoming Bias in Captioning Modelscited by 236
  21. 2020Auditing Differentially Private Machine Learning: How Private is Private SGD?cited by 213
  22. 2018Gazelle: A Low Latency Framework for Secure Neural Network Inferencecited by 193
  23. 2019Towards Fairer Datasets: Filtering and Balancing the Distribution of the People Subtree in the ImageNet Hierarchycited by 156
  24. 2017No Classification without Representation: Assessing Geodiversity Issues in Open Data Sets for the Developing Worldcited by 152
  25. 2014Deep Convolutional Neural Fields for Depth Estimation from a Single Imagecited by 106
  26. 2018Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardwarecited by 101
  27. 2019Gmail Smart Compose: Real-Time Assisted Writingcited by 97
  28. 2019Membership Inference Attacks on Sequence-to-Sequence Models: Is My Data In Your Machine Translation System?cited by 68
  29. 2018Learning to Anonymize Faces for Privacy Preserving Action Detectioncited by 63
  30. 2017Natural and Effective Obfuscation by Head Inpaintingcited by 60
  31. 2016Examining the Impact of Blur on Recognition by Convolutional Networkscited by 58
  32. 2018Low Latency Privacy Preserving Inferencecited by 57
  33. 2018Towards Privacy-Preserving Visual Recognition via Adversarial Training: A Pilot Studycited by 48
  34. 2016Privacy-Preserving Human Activity Recognition from Extreme Low Resolutioncited by 43
  35. 2016Defeating Image Obfuscation with Deep Learningcited by 43
  36. 2017Adversarial Image Perturbation for Privacy Protection -- A Game Theory Perspectivecited by 38
  37. 2016Faceless Person Recognition; Privacy Implications in Social Mediacited by 27
  38. 2017Connecting Pixels to Privacy and Utility: Automatic Redaction of Private Information in Imagescited by 24

50 most cited of 59 All 59 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 14; 2022: 25; 2023: 23; 2024: 14; 2025: 6; 2026: 1.

83 papers.

Sort:

By area:

  1. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  2. 2022DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Stepscited by 1,961
  3. 2023DataComp: In search of the next generation of multimodal datasetscited by 635
  4. 2021Intriguing Properties of Vision Transformerscited by 446
  5. 2022Generalized Category Discoverycited by 209
  6. 2021RedCaps: web-curated image-text data created by the people, for the peoplecited by 175
  7. 2023Emu: Generative Pretraining in Multimodalitycited by 156
  8. 2022Natural Language Descriptions of Deep Visual Featurescited by 136
  9. 2021Understanding and Evaluating Racial Biases in Image Captioningcited by 108
  10. 2024A Survey of Multimodal Large Language Model from A Data-centric Perspectivecited by 73
  11. 2021On Improving Adversarial Transferability of Vision Transformerscited by 70
  12. 2023Training-free Linear Image Inverses via Flowscited by 70
38 more
  1. 2021Privacy-Preserving Portrait Mattingcited by 60
  2. 2024From Pixels to Prose: A Large Dataset of Dense Image Captionscited by 52
  3. 2021PASS: An ImageNet replacement for self-supervised pretraining without humanscited by 51
  4. 2023GeoDE: a Geographically Diverse Evaluation Dataset for Object Recognitioncited by 50
  5. 2021Mitigating Dataset Harms Requires Stewardship: Lessons from 1000 Paperscited by 49
  6. 2024Towards Privacy-Aware Sign Language Translation at Scalecited by 42
  7. 2024Generative Artificial Intelligence: A Systematic Review and Applicationscited by 37
  8. 2022ModelDiff: A Framework for Comparing Learning Algorithmscited by 30
  9. 2022Rethinking Portrait Matting with Privacy Preservingcited by 25
  10. 2022Guillotine Regularization: Why removing layers is needed to improve generalization in Self-Supervised Learningcited by 25
  11. 2022Bugs in the Data: How ImageNet Misrepresents Biodiversitycited by 24
  12. 2023High-performance deep spiking neural networks with 0.3 spikes per neuroncited by 22
  13. 2022Bamboo: Building Mega-Scale Vision Dataset Continually with Human-Machine Synergycited by 20
  14. 2023Visual Atoms: Pre-training Vision Transformers with Sinusoidal Wavescited by 20
  15. 2021A Framework for Deprecating Datasets: Standardizing Documentation, Identification, and Communicationcited by 18
  16. 2022Improving Federated Learning Face Recognition via Privacy-Agnostic Clusterscited by 18
  17. 2023Ethical Considerations for Responsible Data Curationcited by 17
  18. 2025The Curse of Conditions: Analyzing and Improving Optimal Transport for Conditional Flow-Based Generationcited by 16
  19. 2022Double Bubble, Toil and Trouble: Enhancing Certified Robustness through Transitivitycited by 15
  20. 2023Objectives Matter: Understanding the Impact of Self-Supervised Objectives on Vision Transformer Representationscited by 15
  21. 2024A Versatile Framework for Multi-scene Person Re-identificationcited by 14
  22. 2023TMI! Finetuned Models Leak Private Information from their Pretraining Datacited by 13
  23. 2021Learning Representations with Contrastive Self-Supervised Learning for Histopathology Applicationscited by 12
  24. 2022An Exact Mapping From ReLU Networks to Spiking Neural Networkscited by 12
  25. 2023Does Image Anonymization Impact Computer Vision Training?cited by 12
  26. 2022CARETS: A Consistency And Robustness Evaluative Test Suite for VQAcited by 11
  27. 2023DMLR: Data-centric Machine Learning Research -- Past, Present and Futurecited by 11
  28. 2024Making Better Use of Unlabelled Data in Bayesian Active Learningcited by 10
  29. 2023GANonymization: A GAN-based Face Anonymization Framework for Preserving Emotional Expressionscited by 9
  30. 2023SegRCDB: Semantic Segmentation via Formula-Driven Supervised Learningcited by 8
  31. 2024D'ej`a Vu Memorization in Vision-Language Modelscited by 8
  32. 2023Manipulating Transfer Learning for Property Inferencecited by 7
  33. 2022Learnable Privacy-Preserving Anonymization for Pedestrian Imagescited by 6
  34. 2022Progressive Cross-modal Knowledge Distillation for Human Action Recognitioncited by 6
  35. 2023A View From Somewhere: Human-Centric Face Representationscited by 6
  36. 2021Anonymization for Skeleton Action Recognitioncited by 5
  37. 2022Regulating Facial Processing Technologies: Tensions Between Legal and Technical Considerations in the Application of Illinois BIPAcited by 5
  38. 2022Auditing Visualizations: Transparency Methods Struggle to Detect Anomalous Behaviorcited by 5

50 most cited of 83 All 83 (JSON)