The Influence of Faulty Labels in Data Sets on Human Pose Estimation

• Published in: arXiv.org
• Main Authors: Schwarz, Arnold, Hernadi, Levente, Bießmann, Felix, Hildebrand, Kristian
• Format: Paper
• Language: English
• Published: Ithaca Cornell University Library, arXiv.org 09.09.2024
• Subjects: Datasets; Human performance; Impact analysis; Labels; Performance measurement; Pose estimation

In this study we provide empirical evidence demonstrating that the quality of training data impacts model performance in Human Pose Estimation (HPE). Inaccurate labels in widely used data sets, ranging from minor errors to severe mislabeling, can negatively influence learning and distort performance...