mlsanityData to 7 Oct 2026

Predictive Inequity in Object Detection

arXiv 1902.11097 ↗First posted 21 Feb 2019cs.CV cs.LG stat.MLcited by 77

Not on the map. Why?

Abstract (arXiv)

In this work, we investigate whether state-of-the-art object detection systems have equitable predictive performance on pedestrians with different skin tones. This work is motivated by many recent examples of ML and vision systems displaying higher error rates for certain demographic groups than others. We annotate an existing large scale dataset which contains pedestrians, BDD100K, with Fitzpatrick skin tones in ranges [1-3] or [4-6]. We then provide an in-depth comparative analysis of performance between these two skin tone groupings, finding that neither time of day nor occlusion explain this behavior, suggesting this disparity is not merely the result of pedestrians in the 4-6 range appearing in more difficult scenes for detection. We investigate to what extent time of day, occlusion, and reweighting the supervised loss during training affect this predictive bias.

Key facts. Cited by 77 papers in the map (data through 7 Oct 2026). First posted on arXiv 21 Feb 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 14; 2022: 20; 2023: 25; 2024: 11; 2025: 6; 2026: 1.

77 papers.

Sort:

By area:

  1. 2021Understanding and Evaluating Racial Biases in Image Captioningcited by 108
  2. 2022A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learningcited by 104
  3. 2021A Survey on Bias in Visual Datasetscited by 84
  4. 2023FACET: Fairness in Computer Vision Evaluation Benchmarkcited by 58
  5. 2023Uncurated Image-Text Datasets: Shedding Light on Demographic Biascited by 51
  6. 2021A Step Toward More Inclusive People Annotations for Fairnesscited by 46
  7. 2022American == White in Multimodal Language-and-Image AIcited by 39
  8. 2023T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image Generationcited by 39
  9. 2022Quantifying Societal Bias Amplification in Image Captioningcited by 37
  10. 2022Measuring Representational Harms in Image Captioningcited by 36
  11. 2022Towards Racially Unbiased Skin Tone Estimation via Scene Disambiguationcited by 32
  12. 2022Evidence for Hypodescent in Visual Semantic AIcited by 30
38 more
  1. 2022Markedness in Visual Semantic AIcited by 29
  2. 2023Survey of Social Bias in Vision-Language Modelscited by 26
  3. 2023FACTS: First Amplify Correlations and Then Slice to Discover Biascited by 26
  4. 2021Measuring Model Biases in the Absence of Ground Truthcited by 24
  5. 2021UDIS: Unsupervised Discovery of Bias in Deep Visual Recognition Modelscited by 23
  6. 2022Men Also Do Laundry: Multi-Attribute Bias Amplificationcited by 21
  7. 2024LLM-Driven Robots Risk Enacting Discrimination, Violence, and Unlawful Actionscited by 21
  8. 2021Latent Space Smoothing for Individually Fair Representationscited by 18
  9. 2022Symphony: Composing Interactive Interfaces for Machine Learningcited by 18
  10. 2023Ethical Considerations for Responsible Data Curationcited by 17
  11. 2023Model-Agnostic Gender Debiased Image Captioningcited by 17
  12. 2023Beyond Skin Tone: A Multidimensional Measure of Apparent Skin Colorcited by 17
  13. 2021Can Explainable AI Explain Unfairness? A Framework for Evaluating Explainable AIcited by 16
  14. 2021Reliability and Validity of Image-Based and Self-Reported Skin Phenotype Metricscited by 15
  15. 2024Who's in and who's out? A case study of multimodal CLIP-filtering in DataCompcited by 15
  16. 2022Robots Enact Malignant Stereotypescited by 14
  17. 2022MMD-B-Fair: Learning Fair Representations with Statistical Testingcited by 13
  18. 2024Fairness and Bias Mitigation in Computer Vision: A Surveycited by 12
  19. 2021From Learning to Relearning: A Framework for Diminishing Bias in Social Robot Navigationcited by 11
  20. 2021Does Data Repair Lead to Fair Models? Curating Contextually Fair Data To Reduce Model Biascited by 11
  21. 2022Speciesist bias in AI -- How AI applications perpetuate discrimination and unfair outcomes against animalscited by 11
  22. 2022How Robust is Your Fairness? Evaluating and Sustaining Fairness under Unseen Distribution Shiftscited by 11
  23. 2022Simplicity Bias Leads to Amplified Performance Disparitiescited by 11
  24. 2023Skin Deep: Investigating Subjectivity in Skin Tone Annotations for Computer Vision Benchmark Datasetscited by 11
  25. 2024A Unified Framework and Dataset for Assessing Societal Bias in Vision-Language Modelscited by 11
  26. 2022Fairness and Bias in Robot Learningcited by 10
  27. 2023Bias Behind the Wheel: Fairness Testing of Autonomous Driving Systemscited by 7
  28. 2023Fairness-aware Vision Transformer via Debiased Self-Attentioncited by 6
  29. 2021DeepFN: Towards Generalizable Facial Action Unit Recognition with Deep Face Normalizationcited by 5
  30. 2022Casual Conversations v2: Designing a large consent-driven dataset to measure algorithmic bias and robustnesscited by 5
  31. 2023(Ir)rationality in AI: State of the Art, Research Challenges and Open Questionscited by 5
  32. 2025ALBAR: Adversarial Learning approach to mitigate Biases in Action Recognitioncited by 5
  33. 2025VISLIX: An XAI Framework for Validating Vision Models with Slice Discovery and Analysiscited by 5
  34. 2021Adaptive Autonomy in Human-on-the-Loop Vision-Based Robotics Systemscited by 4
  35. 2023Mitigating Bias in Visual Transformers via Targeted Alignmentcited by 4
  36. 2023Fairness in Ranking under Disparate Uncertaintycited by 4
  37. 2024PathwayBench: Assessing Routability of Pedestrian Pathway Networks Inferred from Multi-City Imagerycited by 4
  38. 2021Human Imperceptible Attacks and Applications to Improve Fairnesscited by 3

50 most cited of 77 All 77 (JSON)