Hugging Face's logo

Datasets:
Hgodwarrior
/
Huhb3D-Flange-Topology

Tasks:
Other
Image Segmentation
Robotics
Size:
n<1K
Tags:
topology
3d-object
step
cad
industrial
grasp-planning
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
source_file: string
object_bounds: list<item: double>
  child 0, item: double
best_grasp_strategy: struct<method: string, gripper_type: string, target_face_id: int64, approach_direction: list<item: d (... 46 chars omitted)
  child 0, method: string
  child 1, gripper_type: string
  child 2, target_face_id: int64
  child 3, approach_direction: list<item: double>
      child 0, item: double
  child 4, confidence: double
  child 5, reasoning: string
face_recommendations: list<item: struct<face_id: int64, category_name: string, geom_type: string, area: double, grasp_meth (... 142 chars omitted)
  child 0, item: struct<face_id: int64, category_name: string, geom_type: string, area: double, grasp_method: string, (... 130 chars omitted)
      child 0, face_id: int64
      child 1, category_name: string
      child 2, geom_type: string
      child 3, area: double
      child 4, grasp_method: string
      child 5, gripper_type: string
      child 6, approach_direction: list<item: double>
          child 0, item: double
      child 7, grasp_point: list<item: double>
          child 0, item: double
      child 8, confidence: double
      child 9, notes: string
grasp_method_summary: struct<contour_grip: struct<face_count: int64, total_area: double>, surface_grip: struct<face_count: (... 90 chars omitted)
  child 0, contour_grip: struct<face_count: int64, total_area: double>
      child 0, face_count: int64
      child 1, total_area: double
  child 1, surface_grip: struct<face_count: int6
...
t64
industry: string
source: struct<topology_data: string, step_files: string, parser: string>
  child 0, topology_data: string
  child 1, step_files: string
  child 2, parser: string
objects: list<item: struct<name: string, num_faces: int64, num_triangles: int64, shape_bounds: list<item: dou (... 215 chars omitted)
  child 0, item: struct<name: string, num_faces: int64, num_triangles: int64, shape_bounds: list<item: double>, topol (... 203 chars omitted)
      child 0, name: string
      child 1, num_faces: int64
      child 2, num_triangles: int64
      child 3, shape_bounds: list<item: double>
          child 0, item: double
      child 4, topology_categories: list<item: string>
          child 0, item: string
      child 5, category_distribution: struct<FreeSurface: int64, LateralPlane_Z: int64, Boss: int64, ConvexFeature_Bolt: int64>
          child 0, FreeSurface: int64
          child 1, LateralPlane_Z: int64
          child 2, Boss: int64
          child 3, ConvexFeature_Bolt: int64
      child 6, has_step_file: bool
      child 7, has_grasp_recommendations: bool
description: string
display_name: string
legal: struct<dataset_license: string, model_license: string, original_work: bool, commercial_use_allowed:  (... 47 chars omitted)
  child 0, dataset_license: string
  child 1, model_license: string
  child 2, original_work: bool
  child 3, commercial_use_allowed: bool
  child 4, attribution_required: bool
  child 5, note: string
total_faces: int64
generated_at: string
to
{'dataset_name': Value('string'), 'version': Value('string'), 'display_name': Value('string'), 'description': Value('string'), 'industry': Value('string'), 'generated_at': Value('string'), 'num_objects': Value('int64'), 'total_faces': Value('int64'), 'total_triangles': Value('int64'), 'topology_categories_present': List(Value('string')), 'objects': List({'name': Value('string'), 'num_faces': Value('int64'), 'num_triangles': Value('int64'), 'shape_bounds': List(Value('float64')), 'topology_categories': List(Value('string')), 'category_distribution': {'FreeSurface': Value('int64'), 'LateralPlane_Z': Value('int64'), 'Boss': Value('int64'), 'ConvexFeature_Bolt': Value('int64')}, 'has_step_file': Value('bool'), 'has_grasp_recommendations': Value('bool')}), 'total_statistics': {'total_faces': Value('int64'), 'total_triangles': Value('int64'), 'total_category_distribution': {'FreeSurface': Value('int64'), 'LateralPlane_Z': Value('int64'), 'Boss': Value('int64'), 'ConvexFeature_Bolt': Value('int64')}}, 'pricing': {'personal': Value('int64'), 'commercial': Value('int64'), 'currency': Value('string')}, 'legal': {'dataset_license': Value('string'), 'model_license': Value('string'), 'original_work': Value('bool'), 'commercial_use_allowed': Value('bool'), 'attribution_required': Value('bool'), 'note': Value('string')}, 'source': {'topology_data': Value('string'), 'step_files': Value('string'), 'parser': Value('string')}}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 99, in get_rows_or_raise
                  return get_rows(
                         ^^^^^^^^^
                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 77, in get_rows
                  rows_plus_one = list(itertools.islice(ds, rows_max_number + 1))
                                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2815, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2352, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2377, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.12/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/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 310, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 130, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2369, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              source_file: string
              object_bounds: list<item: double>
                child 0, item: double
              best_grasp_strategy: struct<method: string, gripper_type: string, target_face_id: int64, approach_direction: list<item: d (... 46 chars omitted)
                child 0, method: string
                child 1, gripper_type: string
                child 2, target_face_id: int64
                child 3, approach_direction: list<item: double>
                    child 0, item: double
                child 4, confidence: double
                child 5, reasoning: string
              face_recommendations: list<item: struct<face_id: int64, category_name: string, geom_type: string, area: double, grasp_meth (... 142 chars omitted)
                child 0, item: struct<face_id: int64, category_name: string, geom_type: string, area: double, grasp_method: string, (... 130 chars omitted)
                    child 0, face_id: int64
                    child 1, category_name: string
                    child 2, geom_type: string
                    child 3, area: double
                    child 4, grasp_method: string
                    child 5, gripper_type: string
                    child 6, approach_direction: list<item: double>
                        child 0, item: double
                    child 7, grasp_point: list<item: double>
                        child 0, item: double
                    child 8, confidence: double
                    child 9, notes: string
              grasp_method_summary: struct<contour_grip: struct<face_count: int64, total_area: double>, surface_grip: struct<face_count: (... 90 chars omitted)
                child 0, contour_grip: struct<face_count: int64, total_area: double>
                    child 0, face_count: int64
                    child 1, total_area: double
                child 1, surface_grip: struct<face_count: int6
              ...
              t64
              industry: string
              source: struct<topology_data: string, step_files: string, parser: string>
                child 0, topology_data: string
                child 1, step_files: string
                child 2, parser: string
              objects: list<item: struct<name: string, num_faces: int64, num_triangles: int64, shape_bounds: list<item: dou (... 215 chars omitted)
                child 0, item: struct<name: string, num_faces: int64, num_triangles: int64, shape_bounds: list<item: double>, topol (... 203 chars omitted)
                    child 0, name: string
                    child 1, num_faces: int64
                    child 2, num_triangles: int64
                    child 3, shape_bounds: list<item: double>
                        child 0, item: double
                    child 4, topology_categories: list<item: string>
                        child 0, item: string
                    child 5, category_distribution: struct<FreeSurface: int64, LateralPlane_Z: int64, Boss: int64, ConvexFeature_Bolt: int64>
                        child 0, FreeSurface: int64
                        child 1, LateralPlane_Z: int64
                        child 2, Boss: int64
                        child 3, ConvexFeature_Bolt: int64
                    child 6, has_step_file: bool
                    child 7, has_grasp_recommendations: bool
              description: string
              display_name: string
              legal: struct<dataset_license: string, model_license: string, original_work: bool, commercial_use_allowed:  (... 47 chars omitted)
                child 0, dataset_license: string
                child 1, model_license: string
                child 2, original_work: bool
                child 3, commercial_use_allowed: bool
                child 4, attribution_required: bool
                child 5, note: string
              total_faces: int64
              generated_at: string
              to
              {'dataset_name': Value('string'), 'version': Value('string'), 'display_name': Value('string'), 'description': Value('string'), 'industry': Value('string'), 'generated_at': Value('string'), 'num_objects': Value('int64'), 'total_faces': Value('int64'), 'total_triangles': Value('int64'), 'topology_categories_present': List(Value('string')), 'objects': List({'name': Value('string'), 'num_faces': Value('int64'), 'num_triangles': Value('int64'), 'shape_bounds': List(Value('float64')), 'topology_categories': List(Value('string')), 'category_distribution': {'FreeSurface': Value('int64'), 'LateralPlane_Z': Value('int64'), 'Boss': Value('int64'), 'ConvexFeature_Bolt': Value('int64')}, 'has_step_file': Value('bool'), 'has_grasp_recommendations': Value('bool')}), 'total_statistics': {'total_faces': Value('int64'), 'total_triangles': Value('int64'), 'total_category_distribution': {'FreeSurface': Value('int64'), 'LateralPlane_Z': Value('int64'), 'Boss': Value('int64'), 'ConvexFeature_Bolt': Value('int64')}}, 'pricing': {'personal': Value('int64'), 'commercial': Value('int64'), 'currency': Value('string')}, 'legal': {'dataset_license': Value('string'), 'model_license': Value('string'), 'original_work': Value('bool'), 'commercial_use_allowed': Value('bool'), 'attribution_required': Value('bool'), 'note': Value('string')}, 'source': {'topology_data': Value('string'), 'step_files': Value('string'), 'parser': 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.

Huhb3D Flange Topology Dataset

STEP topology annotated dataset for 4 flange variants (DIN-style) with per-face semantic labels. Includes HorizontalPlane, LateralPlane, Boss, and ConvexFeature labels for robotic grasp planning.

What Makes This Dataset Unique

This is the only publicly available dataset that provides STEP-parsed per-face topology labels for industrial CAD models. Unlike mesh-only datasets, our topology annotations are derived directly from the STEP B-Rep structure using OpenCascade, providing:

  • Per-face semantic labels: Each mesh face is classified into one of 15 topology categories
  • STEP source files: Original parametric CAD models for precise geometry queries
  • Robotic grasp planning: Topology labels enable grasp strategy selection by face type
  • 6DoF pose estimation: Face-level annotations support pose refinement algorithms

Overview

  • 4 industrial mechanical parts
  • 39 total B-Rep faces
  • 10356 total mesh triangles
  • 4 topology categories with per-face semantic labels
  • Industry: Pipe & Flange Manufacturing
  • Source models: STEP (ISO 10303-21) CAD files

Directory Structure 

Huhb3D-Flange-Topology/
  README.md
  LICENSE
  DATASET_METADATA.json
  checksums.sha256
  source_step/
    flange.step
    ...
  objects/
    flange/
      topology_labels.json       # Per-triangle topology labels
      topology_summary.json      # Topology statistics summary
      grasp_recommendations.json # Grasp recommendations (if available)
    ...

Object List

Object Faces Triangles STEP Grasp Topology Categories
flange 8 2516 ✓ ✓ Boss, ConvexFeature_Bolt, FreeSurface, LateralPlane_Z
flange_small 13 2028 ✓ ✓ Boss, ConvexFeature_Bolt, FreeSurface, LateralPlane_Z
flange_medium 8 2516 ✓ ✓ Boss, ConvexFeature_Bolt, FreeSurface, LateralPlane_Z
flange_large 10 3296 ✓ ✓ Boss, ConvexFeature_Bolt, FreeSurface, LateralPlane_Z

Topology Categories

ID Category Description Color
0 FreeSurface 自由曲面（圆柱面、圆锥面、B样条曲面等） #7F7F7F
3 LateralPlane_Z 法线平行于 Z 轴的竖直平面 #FF0000
8 ConvexFeature_Bolt 凸起圆柱特征（螺栓凸台、销钉、轴段） #7F00FF
11 Boss 凸起圆柱台（安装凸台、垫台） #00CC66

Grasp Recommendations

Some objects include grasp_recommendations.json with pre-computed robotic grasp poses. These files are generated by the grasp_recommendations.py module and contain:

  • Recommended grasp approach directions
  • Gripper opening widths
  • Grasp quality scores

Data Format

topology_labels.json

Per-object file containing:

  • source_file: Original STEP file name
  • total_triangles: Total number of mesh triangles
  • total_faces: Total number of B-Rep faces
  • shape_bounds: Bounding box [xmin, ymin, zmin, xmax, ymax, zmax]
  • category_names: Mapping from category ID to name
  • triangle_labels: Array of category IDs, one per triangle
  • faces: Array of face objects with:
    • face_id, geom_type, category_id, category_name
    • area, triangle_count, triangle_start
    • extra: Optional dict with radius, axis_direction, normal

topology_summary.json

Per-object file containing:

  • source_file, total_faces, total_triangles, shape_bounds
  • categories: Dict mapping category ID to name, face_count, triangle_count, total_area

grasp_recommendations.json (optional)

Per-object file containing pre-computed grasp poses with approach directions, gripper widths, and quality scores.

Pricing

License Price
Personal / Academic $29
Commercial $79

All prices in USD. Commercial license permits use in proprietary products.

Citation

If you use this dataset in your research, please cite:

@dataset{huhb3d_flange_topology,
  title   = {Huhb3D Flange Topology Dataset},
  author  = {Huhb},
  year    = {2026},
  version = {1.0.0},
  url     = {https://github.com/huhb-ai/Huhb3D-Topology-Dataset}
}

License

  • 3D Models: CC0 (Public Domain) — original creations, no restrictions
  • Dataset (annotations, metadata, packaging): CC-BY-4.0 — attribution required

You are free to share and adapt for any purpose, including commercially, as long as appropriate credit is given.

See LICENSE for the full license text.

Contact

Downloads last month
7
Total file size:
299 kB