mlsanityData to 14 Oct 2026

A Step Toward More Inclusive People Annotations for Fairness

arXiv 2105.02317 ↗First posted 5 May 2021cs.CVcited by 46

Not on the map. Why?

Abstract (arXiv)

The Open Images Dataset contains approximately 9 million images and is a widely accepted dataset for computer vision research. As is common practice for large datasets, the annotations are not exhaustive, with bounding boxes and attribute labels for only a subset of the classes in each image. In this paper, we present a new set of annotations on a subset of the Open Images dataset called the MIAP (More Inclusive Annotations for People) subset, containing bounding boxes and attributes for all of the people visible in those images. The attributes and labeling methodology for the MIAP subset were designed to enable research into model fairness. In addition, we analyze the original annotation methodology for the person class and its subclasses, discussing the resulting patterns in order to inform future annotation efforts. By considering both the original and exhaustive annotation sets, researchers can also now study how systematic patterns in training annotations affect modeling.

Key facts. Cited by 46 papers in the map (data through 14 Oct 2026). Builds on 16 of its 33 references that are in the map. First posted on arXiv 5 May 2021.

Builds on

16 of 33 references.

References · 16

  1. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  2. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  3. 2014Deep Learning Face Attributes in the Wildcited by 3,693
  4. 2016Equality of Opportunity in Supervised Learningcited by 2,007
  5. 2018Model Cards for Model Reportingcited by 1,080
  6. 2017Age Progression/Regression by Conditional Adversarial Autoencodercited by 429
  7. 2018Learning Adversarially Fair and Transferable Representationscited by 353
  8. 2019Towards Fairer Datasets: Filtering and Balancing the Distribution of the People Subtree in the ImageNet Hierarchycited by 156
  9. 2017No Classification without Representation: Assessing Geodiversity Issues in Open Data Sets for the Developing Worldcited by 153
  10. 2019Does Object Recognition Work for Everyone?cited by 151
  11. 2018Turning a Blind Eye: Explicit Removal of Biases and Variation from Deep Neural Network Embeddingscited by 114
  12. 2019Diversity in Facescited by 77
4 more
  1. 2019Predictive Inequity in Object Detectioncited by 77
  2. 2019Putting Fairness Principles into Practice: Challenges, Metrics, and Improvementscited by 70
  3. 2019Auditing ImageNet: Towards a Model-driven Framework for Annotating Demographic Attributes of Large-Scale Image Datasetscited by 25
  4. 2018Soft Sampling for Robust Object Detectioncited by 22

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 2; 2022: 7; 2023: 21; 2024: 13; 2025: 1; 2026: 2.

46 papers.

Sort:

By area:

  1. 2023Segment Anythingcited by 8,234
  2. 2022PaLI: A Jointly-Scaled Multilingual Language-Image Modelcited by 766
  3. 2023VideoPoet: A Large Language Model for Zero-Shot Video Generationcited by 492
  4. 2023PaLI-X: On Scaling up a Multilingual Vision and Language Modelcited by 255
  5. 2023PaLI-3 Vision Language Models: Smaller, Faster, Strongercited by 126
  6. 2023FACET: Fairness in Computer Vision Evaluation Benchmarkcited by 58
  7. 2023Adversarial attacks and defenses in explainable artificial intelligence: A surveycited by 48
  8. 2022Gender Artifacts in Visual Datasetscited by 38
  9. 2024CLIP the Bias: How Useful is Balancing Data in Multimodal Learning?cited by 35
  10. 2023Overwriting Pretrained Bias with Finetuning Datacited by 32
  11. 2023Survey of Social Bias in Vision-Language Modelscited by 26
  12. 2022Fairness Indicators for Systematic Assessments of Visual Feature Extractorscited by 24
34 more
  1. 2023Auditing Gender Presentation Differences in Text-to-Image Modelscited by 22
  2. 2023Consensus and Subjectivity of Skin Tone Annotation for ML Fairnesscited by 22
  3. 2024Holistic Safety and Responsibility Evaluations of Advanced AI Modelscited by 19
  4. 2023Ethical Considerations for Responsible Data Curationcited by 17
  5. 2023Quantifying Bias in Text-to-Image Generative Modelscited by 16
  6. 2024Revealing and Reducing Gender Biases in Vision and Language Assistants (VLAs)cited by 15
  7. 2023Evaluating the Fairness of Discriminative Foundation Models in Computer Visioncited by 12
  8. 2021Robustness Disparities in Commercial Face Detectioncited by 11
  9. 2023Ground Truth Or Dare: Factors Affecting The Creation Of Medical Datasets For Training AIcited by 11
  10. 2023DIG In: Evaluating Disparities in Image Generations with Indicators for Geographic Diversitycited by 9
  11. 2024Resampled Datasets Are Not Enough: Mitigating Societal Bias Beyond Single Attributescited by 7
  12. 2022Measuring and signing fairness as performance under multiple stakeholder distributionscited by 6
  13. 2023A View From Somewhere: Human-Centric Face Representationscited by 6
  14. 2024Generalized People Diversity: Learning a Human Perception-Aligned Diversity Representation for People Imagescited by 6
  15. 2024Copycats: the many lives of a publicly available medical imaging datasetcited by 6
  16. 2026Interpretable Debiasing of Vision-Language Models for Social Fairnesscited by 6
  17. 2022Robustness Disparities in Face Detectioncited by 5
  18. 2023Towards Reliable Assessments of Demographic Disparities in Multi-Label Image Classifierscited by 5
  19. 2023BELT:Bootstrapping Electroencephalography-to-Language Decoding and Zero-Shot Sentiment Classification by Natural Language Supervisioncited by 5
  20. 2023DSAP: Analyzing Bias Through Demographic Comparison of Datasetscited by 5
  21. 2024Annotations on a Budget: Leveraging Geo-Data Similarity to Balance Model Performance and Annotation Costcited by 5
  22. 2025Bias in Gender Bias Benchmarks: How Spurious Features Distort Evaluationcited by 5
  23. 2023The Casual Conversations v2 Datasetcited by 4
  24. 2024Measuring Machine Learning Harms from Stereotypes Requires Understanding Who Is Harmed by Which Errors in What Wayscited by 4
  25. 2022Are Commercial Face Detection Models as Biased as Academic Models?cited by 3
  26. 2023Attribute Annotation and Bias Evaluation in Visual Datasets for Autonomous Drivingcited by 3
  27. 2024A Taxonomy of Challenges to Curating Fair Datasetscited by 2
  28. 2024My Body My Choice: Human-Centric Full-Body Anonymizationcited by 2
  29. 2026The Consensus Trap: Dissecting Subjectivity and the "Ground Truth" Illusion in Data Annotationcited by 2
  30. 2024Wake Vision: A Tailored Dataset and Benchmark Suite for TinyML Computer Vision Applicationscited by 1
  31. 2024For a semiotic AI: Bridging computer vision and visual semiotics for computational observation of large scale facial image archivescited by 1
  32. 2021Improving Ethical Outcomes with Machine-in-the-Loop: Broadening Human Understanding of Data Annotationscited by 0
  33. 2022Addressing Bias in Face Detectors using Decentralised Data collection with incentivescited by 0
  34. 2024Situated Ground Truths: Enhancing Bias-Aware AI by Situating Data Labels with SituAnnotatecited by 0