# Ambiguous Annotations: When is a Pedestrian not a Pedestrian?

arXiv: https://arxiv.org/abs/2405.08794 · HTML: https://mlsanity.com/paper/2405.08794/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 30 references that are in the map.
- First posted on arXiv 14 May 2024.

## Builds on (most cited first, 22 of 22)

- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Deep High-Resolution Representation Learning for Visual Recognition](https://mlsanity.com/paper/1908.07919/) (2019)
- [Cascade R-CNN: Delving into High Quality Object Detection](https://mlsanity.com/paper/1712.00726/) (2017)
- [Learning from Noisy Labels with Deep Neural Networks: A Survey](https://mlsanity.com/paper/2007.08199/) (2020)
- [Confident Learning: Estimating Uncertainty in Dataset Labels](https://mlsanity.com/paper/1911.00068/) (2019)
- [Are we done with ImageNet?](https://mlsanity.com/paper/2006.07159/) (2020)
- [Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotations](https://mlsanity.com/paper/2110.12088/) (2021)
- [Deep learning with noisy labels: exploring techniques and remedies in medical image analysis](https://mlsanity.com/paper/1912.02911/) (2019)
- [What Do Compressed Deep Neural Networks Forget?](https://mlsanity.com/paper/1911.05248/) (2019)
- [Learning From Noisy Labels By Regularized Estimation Of Annotator Confusion](https://mlsanity.com/paper/1902.03680/) (2019)
- [A Survey of Label-noise Representation Learning: Past, Present and Future](https://mlsanity.com/paper/2011.04406/) (2020)
- [A Review and Comparative Study on Probabilistic Object Detection in Autonomous Driving](https://mlsanity.com/paper/2011.10671/) (2020)
- [Deep Label Distribution Learning with Label Ambiguity](https://mlsanity.com/paper/1611.01731/) (2016)
- [Image Classification with Deep Learning in the Presence of Noisy Labels: A Survey](https://mlsanity.com/paper/1912.05170/) (2019)
- [The EuroCity Persons Dataset: A Novel Benchmark for Object Detection](https://mlsanity.com/paper/1805.07193/) (2018)
- [Diverse Image Generation via Self-Conditioned GANs](https://mlsanity.com/paper/2006.10728/) (2020)
- [Training Convolutional Networks with Noisy Labels](https://mlsanity.com/paper/1406.2080/) (2014)
- [Improving Inversion and Generation Diversity in StyleGAN using a Gaussianized Latent Space](https://mlsanity.com/paper/2009.06529/) (2020)
- [Label Noise Types and Their Effects on Deep Learning](https://mlsanity.com/paper/2003.10471/) (2020)
- [Pedestrian Detection: Domain Generalization, CNNs, Transformers and Beyond](https://mlsanity.com/paper/2201.03176/) (2022)
- [Label Distribution Learning via Implicit Distribution Representation](https://mlsanity.com/paper/2209.13824/) (2022)
- [Automated Cleanup of the ImageNet Dataset by Model Consensus, Explainability and Confident Learning](https://mlsanity.com/paper/2103.16324/) (2021)

## Built on by (most cited first, 4 of 4)

- [Understanding Model Calibration -- A gentle introduction and visual exploration of calibration and the expected calibration error (ECE)](https://mlsanity.com/paper/2501.19047/) (2025)
- [AuralSAM2: Enabling SAM2 Hear Through Pyramid Audio-Visual Feature Prompting](https://mlsanity.com/paper/2506.01015/) (2025)
- [Quantifying Ambiguity in Categorical Annotations: A Measure and Statistical Inference Framework](https://mlsanity.com/paper/2510.04366/) (2025)
- [An Assessment of Human vs. Model Uncertainty in Soft-Label Learning and Calibration](https://mlsanity.com/paper/2605.18648/) (2026)

## Abstract (arXiv)

Datasets labelled by human annotators are widely used in the training and testing of machine learning models. In recent years, researchers are increasingly paying attention to label quality. However, it is not always possible to objectively determine whether an assigned label is correct or not. The present work investigates this ambiguity in the annotation of autonomous driving datasets as an important dimension of data quality. Our experiments show that excluding highly ambiguous data from the training improves model performance of a state-of-the-art pedestrian detector in terms of LAMR, precision and F1 score, thereby saving training time and annotation costs. Furthermore, we demonstrate that, in order to safely remove ambiguous instances and ensure the retained representativeness of the training data, an understanding of the properties of the dataset and class under investigation is crucial.