# Noisy Annotations in Semantic Segmentation

Source: [Noisy Annotations in Semantic Segmentation](<https://www.sota2.com/research/paper/68fa77c81e209131cea754f6>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-02T09:41:41.434Z

Authors: Moshe Kimhi, Omer Kerem, Eden Grad, Ehud Rivlin, Chaim Baskin

Published: 2024-06-16T10:49:23.000Z

## Abstract

Obtaining accurate labels for instance segmentation is particularly challenging due to the complex nature of the task. Each image necessitates multiple annotations, encompassing not only the object class but also its precise spatial boundaries. These requirements elevate the likelihood of errors and inconsistencies in both manual and automated annotation processes. By simulating different noise conditions, we provide a realistic scenario for assessing the robustness and generalization capabilities of instance segmentation models in different segmentation tasks, introducing COCO-N and Cityscapes-N. We also propose a benchmark for weakly annotation noise, dubbed COCO-WAN, which utilizes foundation models and weak annotations to simulate semi-automated annotation tools and their noisy labels. This study sheds light on the quality of segmentation masks produced by various models and challenges the efficacy of popular methods designed to address learning with label noise.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2406.10891>)

[PDF](<https://arxiv.org/pdf/2406.10891.pdf>)

[Implementation](<https://github.com/eden500/Noisy-Labels-Instance-Segmentation>)