Datasets:

Totorotibo84
/

pitchlens-v002

Error code:   DatasetGenerationError
Exception:    IndexError
Message:      list index out of range
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1859, in _prepare_split_single
                  original_shard_lengths[original_shard_id] += len(table)
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^
              IndexError: list index out of range
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      builder, max_dataset_size_bytes=max_dataset_size_bytes
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1694, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  ):
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1880, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

text string
0 0.924609 0.431944 0.010156 0.016667
0 0.924609 0.431944 0.010156 0.016667
0 0.924609 0.431944 0.010156 0.016667
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.924609 0.431250 0.010156 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.924609 0.431250 0.010156 0.015278
0 0.924609 0.431944 0.010156 0.016667
0 0.924609 0.431250 0.010156 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431944 0.010937 0.016667
0 0.925000 0.431944 0.010937 0.016667
0 0.925781 0.431250 0.010937 0.015278
0 0.925000 0.431944 0.010937 0.016667
0 0.925000 0.431250 0.010937 0.015278
0 0.925000 0.431944 0.010937 0.016667
0 0.925000 0.431944 0.010937 0.016667
0 0.924609 0.431250 0.010156 0.015278
0 0.924609 0.431250 0.010156 0.015278
0 0.924609 0.431250 0.010156 0.015278
0 0.924609 0.431250 0.010156 0.015278
0 0.924609 0.431250 0.010156 0.015278
0 0.925391 0.431250 0.010156 0.015278
0 0.925391 0.432639 0.010156 0.015278
0 0.925391 0.431944 0.010156 0.016667
0 0.925391 0.431250 0.010156 0.015278
0 0.925391 0.431944 0.010156 0.016667
0 0.925391 0.431944 0.010156 0.016667
0 0.925391 0.431944 0.010156 0.016667
0 0.925391 0.432639 0.010156 0.015278
0 0.925391 0.432639 0.010156 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.925000 0.432639 0.009375 0.015278
0 0.729297 0.847917 0.022656 0.029167
0 0.751953 0.842361 0.021094 0.031944
0 0.767188 0.829861 0.018750 0.031944
0 0.773438 0.826389 0.018750 0.027778
0 0.789844 0.822222 0.015625 0.027778
0 0.803906 0.810417 0.015625 0.026389
0 0.812109 0.804861 0.019531 0.031944
0 0.823438 0.800694 0.015625 0.026389
0 0.810937 0.799306 0.025000 0.026389
0 0.786328 0.790278 0.032031 0.027778
0 0.745313 0.778472 0.028125 0.026389
0 0.700781 0.763194 0.028125 0.023611
0 0.683594 0.754861 0.025000 0.029167
0 0.654297 0.741667 0.021094 0.030556
0 0.631250 0.733333 0.018750 0.025000
0 0.621094 0.727083 0.015625 0.026389
0 0.601562 0.719444 0.018750 0.027778
0 0.582031 0.708333 0.018750 0.019444
0 0.576953 0.709722 0.016406 0.025000
0 0.555859 0.702083 0.011719 0.026389
0 0.536328 0.701389 0.016406 0.022222
0 0.527344 0.699306 0.014063 0.023611
0 0.513672 0.697222 0.014844 0.025000
0 0.508984 0.700000 0.011719 0.019444
0 0.507812 0.699306 0.012500 0.023611
0 0.502734 0.702083 0.011719 0.020833
0 0.499609 0.702083 0.011719 0.020833
0 0.509766 0.700694 0.019531 0.023611
0 0.535156 0.694444 0.021875 0.027778
0 0.558594 0.692361 0.018750 0.023611
0 0.568750 0.690972 0.017188 0.020833
0 0.591016 0.684722 0.019531 0.019444
0 0.607031 0.679167 0.015625 0.022222
0 0.615234 0.679167 0.011719 0.022222
0 0.627734 0.675000 0.013281 0.025000
0 0.638281 0.673611 0.014063 0.022222
0 0.643359 0.673611 0.013281 0.022222
0 0.652734 0.672917 0.011719 0.023611
0 0.659766 0.671528 0.010156 0.026389
0 0.666016 0.673611 0.007031 0.025000

End of preview.

PitchLens Ball Detection v002

PitchLens v002 is a curated football-ball detection dataset built for one practical question: how reliably can a small detector keep the ball in frame for a low-cost automatic match camera?

The release contains 4,549 images and 3,259 bounding boxes in YOLO format. It is derived from the Infactory Soccer Ball Tracking dataset, then audited and corrected for PitchLens. Source sequences remain isolated between train, validation and test.

Dataset summary

Split	Images	Positive images	Negative images	Boxes	Sequences
Train	3,048	2,113	935	2,114	30
Validation	670	494	176	494	7
Test	831	639	192	651	8
Total	4,549	3,246	1,303	3,259	45

There is one class:

0: ball

Empty label files are intentional negative examples.

What changed from the source

The source revision was frozen at commit 75854a5837dd3302cbb1843355ecc1a6f124f818. All 7,152 source frames were reviewed through a resumable annotation interface.

2,603 unusable or ambiguous frames were excluded;
246 bounding boxes were redrawn;
92 frames were explicitly marked as negative;
1,298 source annotations were manually confirmed;
remaining undecided frames retained their source annotation;
exact cross-split duplicates and sequence leakage were checked: none remain.

The repository includes the frozen review decisions, source manifest snapshot, output hashes and build report so that the curation history remains inspectable.

Layout

pitchlens-v002/
├── train/{images,labels}
├── val/{images,labels}
├── test/{images,labels}
├── dataset.yaml
├── manifest.csv
├── parent_manifest_snapshot.csv
├── review_decisions_snapshot.csv
└── build_report.json

Download and use

from huggingface_hub import snapshot_download

dataset_root = snapshot_download(
    repo_id="Totorotibo84/pitchlens-v002",
    repo_type="dataset",
)
print(dataset_root)

For Ultralytics, either run from the downloaded repository or update the path field in dataset.yaml to the local snapshot directory.

Intended use

This release is intended for non-commercial research on football-ball detection, tiny-object detection, tracking and automatic sports-camera prototypes. The validation split is suitable for model selection; keep the test split untouched until the final evaluation.

The images do not yet represent the final PitchLens deployment domain. Most footage comes from existing match video rather than a fixed iPhone on an amateur touchline. Expect domain shift in camera height, compression, lighting, motion blur and ball size.

License and attribution

The upstream dataset is licensed under CC BY-NC 4.0. This derived release uses the same license and is not authorized for commercial use.

Please attribute both:

Infactory Soccer Ball Tracking, the source dataset;
PitchLens v002, for the review, corrections, exclusions and sequence-safe release.

Related model

The matching high-resolution annotation teacher is available at Totorotibo84/pitchlens-yolo26x-v002-teacher.

Downloads last month: 88

Total file size:

937 MB

Models trained or fine-tuned on Totorotibo84/pitchlens-v002

Totorotibo84/pitchlens-yolo26x-v002-teacher

Object Detection • Updated Sep 25 • 81