Datasets:

Kyungwan
/

3dgs-svraster-demo

Tasks:

Other

Tags:

License:

Dataset card Files Files and versions

xet

Community

Dataset Viewer

Duplicate

Split (1)

train

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
schemaVersion: int64
scene: string
coordinateFrame: string
models: list<item: struct<name: string, modelToColmap: list<item: double>>>
  child 0, item: struct<name: string, modelToColmap: list<item: double>>
      child 0, name: string
      child 1, modelToColmap: list<item: double>
          child 0, item: double
referenceCameras: list<item: struct<id: string, R: list<item: double>, t: list<item: double>, fx: double, fy: double,  (... 44 chars omitted)
  child 0, item: struct<id: string, R: list<item: double>, t: list<item: double>, fx: double, fy: double, cx: double, (... 32 chars omitted)
      child 0, id: string
      child 1, R: list<item: double>
          child 0, item: double
      child 2, t: list<item: double>
          child 0, item: double
      child 3, fx: double
      child 4, fy: double
      child 5, cx: double
      child 6, cy: double
      child 7, w: int64
      child 8, h: int64
defaultScene: string
scenes: list<item: struct<id: string, label: string, source: string, assets: struct<3dgs: struct<path: strin (... 178 chars omitted)
  child 0, item: struct<id: string, label: string, source: string, assets: struct<3dgs: struct<path: string, format:  (... 166 chars omitted)
      child 0, id: string
      child 1, label: string
      child 2, source: string
      child 3, assets: struct<3dgs: struct<path: string, format: string, bytes: int64, sha256: string>, svraster: struct<pa (... 107 chars omitted)
          child 0, 3dgs: struct<path: string, format: string, bytes: int64, sha256: string>
              child 0, path: string
              child 1, format: string
              child 2, bytes: int64
              child 3, sha256: string
          child 1, svraster: struct<path: string, format: string, bytes: int64, sha256: string>
              child 0, path: string
              child 1, format: string
              child 2, bytes: int64
              child 3, sha256: string
          child 2, alignment: struct<path: string, format: string>
              child 0, path: string
              child 1, format: string
to
{'schemaVersion': Value('int64'), 'defaultScene': Value('string'), 'scenes': List({'id': Value('string'), 'label': Value('string'), 'source': Value('string'), 'assets': {'3dgs': {'path': Value('string'), 'format': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string')}, 'svraster': {'path': Value('string'), 'format': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string')}, 'alignment': {'path': Value('string'), 'format': Value('string')}}})}
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 483, 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 2840, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2398, 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 2378, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              schemaVersion: int64
              scene: string
              coordinateFrame: string
              models: list<item: struct<name: string, modelToColmap: list<item: double>>>
                child 0, item: struct<name: string, modelToColmap: list<item: double>>
                    child 0, name: string
                    child 1, modelToColmap: list<item: double>
                        child 0, item: double
              referenceCameras: list<item: struct<id: string, R: list<item: double>, t: list<item: double>, fx: double, fy: double,  (... 44 chars omitted)
                child 0, item: struct<id: string, R: list<item: double>, t: list<item: double>, fx: double, fy: double, cx: double, (... 32 chars omitted)
                    child 0, id: string
                    child 1, R: list<item: double>
                        child 0, item: double
                    child 2, t: list<item: double>
                        child 0, item: double
                    child 3, fx: double
                    child 4, fy: double
                    child 5, cx: double
                    child 6, cy: double
                    child 7, w: int64
                    child 8, h: int64
              defaultScene: string
              scenes: list<item: struct<id: string, label: string, source: string, assets: struct<3dgs: struct<path: strin (... 178 chars omitted)
                child 0, item: struct<id: string, label: string, source: string, assets: struct<3dgs: struct<path: string, format:  (... 166 chars omitted)
                    child 0, id: string
                    child 1, label: string
                    child 2, source: string
                    child 3, assets: struct<3dgs: struct<path: string, format: string, bytes: int64, sha256: string>, svraster: struct<pa (... 107 chars omitted)
                        child 0, 3dgs: struct<path: string, format: string, bytes: int64, sha256: string>
                            child 0, path: string
                            child 1, format: string
                            child 2, bytes: int64
                            child 3, sha256: string
                        child 1, svraster: struct<path: string, format: string, bytes: int64, sha256: string>
                            child 0, path: string
                            child 1, format: string
                            child 2, bytes: int64
                            child 3, sha256: string
                        child 2, alignment: struct<path: string, format: string>
                            child 0, path: string
                            child 1, format: string
              to
              {'schemaVersion': Value('int64'), 'defaultScene': Value('string'), 'scenes': List({'id': Value('string'), 'label': Value('string'), 'source': Value('string'), 'assets': {'3dgs': {'path': Value('string'), 'format': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string')}, 'svraster': {'path': Value('string'), 'format': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string')}, 'alignment': {'path': Value('string'), 'format': Value('string')}}})}
              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.

3DGS · SVRaster Demo

Public, browser-ready artifacts for comparing aligned 3D Gaussian Splatting and SVRaster reconstructions. The repository intentionally contains only the two curated demo scenes listed in manifest.json.

Scenes

Scene	Default	3DGS	SVRaster	Runtime alignment
Bicycle	Yes	PLY	WebGL voxel PLY	Yes
Bonsai	No	PLY	WebGL voxel PLY	Yes

Each pair was reconstructed from the same Mip-NeRF 360 scene. The runtime alignment file maps both models into a shared COLMAP world frame and includes one initial camera. It deliberately excludes local paths, training provenance, the complete camera set, validation reports, checkpoints, and training configs.

Layout

manifest.json
scenes/
  bicycle/
    3dgs.ply
    svraster.ply
    alignment.runtime.json
  bonsai/
    3dgs.ply
    svraster.ply
    alignment.runtime.json

Consumers should resolve asset paths relative to manifest.json and pin a Hub commit SHA in production. Files are intentionally large and should be fetched directly from the Hugging Face CDN rather than proxied through an application server.

Source and attribution

The Bicycle and Bonsai scenes originate from the public Mip-NeRF 360 dataset. If you use these artifacts in research, cite the original dataset and the reconstruction methods used by your application.

These artifacts are demo outputs, not original camera images or training checkpoints. No client-provided model is included.

Downloads last month: 155

Total file size:

837 MB