Hugging Face's logo

Datasets:
Dhi-Technologies
/
amodal-counting-benchmark

Tasks:
Object Detection
Size:
n<1K
Tags:
computer-vision
object-counting
occlusion
uncertainty-quantification
calibration
conformal-prediction
License:
Dataset card Files
xet
 Community
Dataset Viewer
Duplicate
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
crowding: double
image_size: list<item: int64>
  child 0, item: int64
detections: list<item: list<item: double>>
  child 0, item: list<item: double>
      child 0, item: double
occluders: list<item: list<item: double>>
  child 0, item: list<item: double>
      child 0, item: double
true_count: int64
detected_count: int64
true_visibilities: list<item: double>
  child 0, item: double
seed: int64
sweep: list<item: struct<crowding: double, curve: struct<k: double, v0: double, support: list<item: double> (... 159 chars omitted)
  child 0, item: struct<crowding: double, curve: struct<k: double, v0: double, support: list<item: double>, n_samples (... 147 chars omitted)
      child 0, crowding: double
      child 1, curve: struct<k: double, v0: double, support: list<item: double>, n_samples: int64, z_scale: double>
          child 0, k: double
          child 1, v0: double
          child 2, support: list<item: double>
              child 0, item: double
          child 3, n_samples: int64
          child 4, z_scale: double
      child 2, naive_mae: double
      child 3, corrected_mae: double
      child 4, ci_coverage: double
      child 5, extrapolation_flagged_rate: double
      child 6, n_eval_scenes: int64
to
{'sweep': List({'crowding': Value('float64'), 'curve': {'k': Value('float64'), 'v0': Value('float64'), 'support': List(Value('float64')), 'n_samples': Value('int64'), 'z_scale': Value('float64')}, 'naive_mae': Value('float64'), 'corrected_mae': Value('float64'), 'ci_coverage': Value('float64'), 'extrapolation_flagged_rate': Value('float64'), 'n_eval_scenes': Value('int64')}), 'seed': Value('int64')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
                  return get_rows(
                      dataset=dataset,
                  ...<4 lines>...
                      column_names=column_names,
                  )
                File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
                  return func(*args, **kwargs)
                File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
                  rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
                File "/src/services/worker/src/worker/utils.py", line 478, in safe_iter
                  yield from ds.decode(False) if ds.features else ds
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2818, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2355, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2380, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2369, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              crowding: double
              image_size: list<item: int64>
                child 0, item: int64
              detections: list<item: list<item: double>>
                child 0, item: list<item: double>
                    child 0, item: double
              occluders: list<item: list<item: double>>
                child 0, item: list<item: double>
                    child 0, item: double
              true_count: int64
              detected_count: int64
              true_visibilities: list<item: double>
                child 0, item: double
              seed: int64
              sweep: list<item: struct<crowding: double, curve: struct<k: double, v0: double, support: list<item: double> (... 159 chars omitted)
                child 0, item: struct<crowding: double, curve: struct<k: double, v0: double, support: list<item: double>, n_samples (... 147 chars omitted)
                    child 0, crowding: double
                    child 1, curve: struct<k: double, v0: double, support: list<item: double>, n_samples: int64, z_scale: double>
                        child 0, k: double
                        child 1, v0: double
                        child 2, support: list<item: double>
                            child 0, item: double
                        child 3, n_samples: int64
                        child 4, z_scale: double
                    child 2, naive_mae: double
                    child 3, corrected_mae: double
                    child 4, ci_coverage: double
                    child 5, extrapolation_flagged_rate: double
                    child 6, n_eval_scenes: int64
              to
              {'sweep': List({'crowding': Value('float64'), 'curve': {'k': Value('float64'), 'v0': Value('float64'), 'support': List(Value('float64')), 'n_samples': Value('int64'), 'z_scale': Value('float64')}, 'naive_mae': Value('float64'), 'corrected_mae': Value('float64'), 'ci_coverage': Value('float64'), 'extrapolation_flagged_rate': Value('float64'), 'n_eval_scenes': Value('int64')}), 'seed': Value('int64')}
              because column names don't match

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.

Amodal Counting Benchmark (A4)

Product: amodal-counting, "count what detectors can't see": visibility-corrected counting through crowds, clutter, and occlusion, with calibrated intervals instead of bare point-certainty.

Synthetic data, real-data validation in progress

Every scene in this dataset is procedurally generated (ground-truth boxes, occlusion, and a simulated detector with a known detectability curve). It is designed so every claim the product makes can be checked against exact ground truth, not to resemble any specific real camera or crowd. Real-world (non-synthetic) validation of these numbers has not been performed yet.

What's in this dataset

  • scenes.jsonl: 160 synthetic scenes (40 per occlusion-density level: crowding in {0.0, 0.3, 0.6, 0.8}), generated by the repo's own amodal.synth.generate_scenes with seed=2 (the exact evaluation population the headline benchmark scores against). Each row:
    {"crowding": 0.0, "image_size": [640, 480],
 "detections": [[x1,y1,x2,y2], ...],
 "occluders": [[x1,y1,x2,y2], ...],
 "true_count": 12, "detected_count": 7,
 "true_visibilities": [1.0, 0.73, ...], "seed": 200000}
    detections are what a simulated real-world detector reports (it misses occluded objects, per a logistic detectability curve); true_count/true_visibilities are ground truth.
  • bench_results.json: the output of amodal.cli bench --seed 0, the fitted detectability curve and naive-vs-corrected accuracy per crowding level.

Measured result (from this repo, reproduced when this dataset was generated)

Naive count = raw detector count. Corrected count = visibility-corrected estimate. Coverage = how often the 90% calibrated interval actually contained the true count.

crowding naive MAE corrected MAE 90% CI coverage
0.0 2.8 2.4 0.90
0.3 2.6 1.97 0.90
0.6 2.85 2.34 0.975
0.8 3.73 2.46 0.975

Reproduce with: PYTHONPATH=src .venv/bin/python -m amodal.cli bench --seed 0

Schema notes

  • Boxes are [x1, y1, x2, y2] in pixels, image origin top-left.
  • true_visibilities is the ground-truth fraction of each true object NOT covered by occluders, other objects, or the frame border: the quantity the estimator has to recover from detections alone.

Method card, no trained weights

This product is pure Python (numpy/scipy) math, not a trained model. There are no weights to download: the "detector" here is a simulated detectability curve, and the correction is Horvitz-Thompson estimation with conformal interval calibration. The honest finding worth flagging: uncapped correction can be worse than naive counting when a few near-zero-probability detections dominate the variance (measured per-scene MAE 2.89 vs 2.63 naive at crowding 0.6 with min_p=0.1), which is why the weight cap is set by measurement, not taste. See the calibration deep-dive post.

Try it

Source & research context

Downloads last month
12
Total file size:
191 kB

Space using Dhi-Technologies/amodal-counting-benchmark 1

Collection including Dhi-Technologies/amodal-counting-benchmark