SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2207.06104 · 2022

Automated Detection of Label Errors in Semantic Segmentation Datasets via Deep Learning and Uncertainty Quantification

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2207.06104.json · badge: /api/ran/2207.06104.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
mrcoee/automatic-label-error-detection canonical (classifier label)
classifier basis not recorded on the link
2 ran · 1 no recorded run
FunctionStatusWhere it lives
metrics_to_dataset Ran
ran on a synthesized input
mrcoee/automatic-label-error-detection/src/analyze_metrics.py
code served (permissive licence) · get_code("3d4cdd412601c250")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create a minimal metrics dictionary that the function expects
metrics = {
    'S_in': [1, 2, 0, 1, 3],  # 5 entries
    'metric1': [0.1, 0.2, 0.3, 0.4, 0.5],
    'metric2': [0.5, 0.6, 0.7, 0.8, 0.9],
    'class': [1, 2, 1, 2, 1],
    'iou': [0.75, 0.80, 0.85, 0.90, 0.95],
    'iou0': [0.70, 0.75, 0.80, 0.85, 0.90],
}

ARGS = {
    'metrics': metrics,
    'x_names': ['metric1', 'metric2'],
    'prob_names': ['class'],
    'non_empty': True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
metrics_to_nparray Ran
ran on a synthesized input
mrcoee/automatic-label-error-detection/src/analyze_metrics.py
code served (permissive licence) · get_code("478af4f4143a9fdb")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create a minimal metrics dictionary
# The function accesses metrics['S_in'] and metrics[name] for each name in names
# S_in is used to filter indices, so it needs values > 0 for non_empty filtering
# Other metrics should be array-like with matching length

metrics = {
    'S_in': np.array([1.0, 2.0, 0.5, 3.0]),  # 4 samples, some > 0 for filtering
    'accuracy': np.array([0.85, 0.90, 0.75, 0.88]),
    'precision': np.array([0.82, 0.89, 0.72, 0.87]),
    'class': np.array([1, 2, 1, 3])
}

# names parameter: list of metric keys to extract
names = ['accuracy', 'precision', 'class']

# normalize parameter: boolean to apply z-score normalization
normalize = True

# non_empty parameter: boolean to filter by S_in > 0
non_empty = True

ARGS = {
    'metrics': metrics,
    'names': names,
    'normalize': normalize,
    'non_empty': non_empty
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (4, 3) — shape (4, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
classification_fit_and_predict Not yet run
unverified (harvested, no recorded run)
mrcoee/automatic-label-error-detection/src/regression.py
code served (permissive licence) · get_code("4d642f9d3d7d7c5c")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

In this work, we for the first time present a method for detecting label errors in image datasets with semantic segmentation, i.e., pixel-wise class labels. Annotation acquisition for semantic segmentation datasets is time-consuming and requires plenty of human labor. In particular, review processes are time consuming and label errors can easily be overlooked by humans. The consequences are biased benchmarks and in extreme cases also performance degradation of deep neural networks (DNNs) trained on such datasets. DNNs for semantic segmentation yield pixel-wise predictions, which makes detection of label errors via uncertainty quantification a complex task. Uncertainty is particularly pronounced at the transitions between connected components of the prediction. By lifting the consideration of uncertainty to the level of predicted components, we enable the usage of DNNs together with component-level uncertainty quantification for the detection of label errors. We present a principled approach to benchmarking the task of label error detection by dropping labels from the Cityscapes dataset as well from a dataset extracted from the CARLA driving simulator, where in the latter case we have the labels under control. Our experiments show that our approach is able to detect the vast majority of label errors while controlling the number of false label error detections. Furthermore, we apply our method to semantic segmentation datasets frequently used by the computer vision community and present a collection of label errors along with sample statistics.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2207.06104")
get_code_for_paper("2207.06104")
have("2207.06104")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2207.06104.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2207.06104.svg)](https://syntology.ai/paper/2207.06104)

Connect an agent — have() is free.