SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2006.09102 · NeurIPS · 2020

UCSG-NET -Unsupervised Discovering of Constructive Solid Geometry Tree

Kacper Kania, Maciej Zięba, Tomasz Kajdanowicz

arXiv · PDF · Open in the Atlas

Code that ran

8 of 10 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 8 have no contract verdict; 0 failures of our instrument; 2 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/2006.09102.json · badge: /api/ran/2006.09102.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
kacperkan/ucsgnet canonical (classifier label)
classifier basis: census
8 ran · 2 no recorded run
FunctionStatusWhere it lives
generate_single_image Ran
ran on a synthesized input
kacperkan/ucsgnet/ucsgnet/generators/generate_cross_diamond_dataset.py
code served (permissive licence) · get_code("dc9c028cc9e9b9f2")
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 Image. 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 PIL.Image
import numpy as np

ARGS = {
    'height': 256,
    'width': 384,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 2 and dtype family int describe that coercion, not the output, and are not 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
get_chamfer_distance_and_normal_consistency Ran
ran on a synthesized input
kacperkan/ucsgnet/ucsgnet/ucsgnet/evaluate_on_3d_data.py
code served (permissive licence) · get_code("88c737ef55663bd3")
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, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 torch

torch.manual_seed(0)

# Create small tensors for ground truth points
gt_points = torch.randn(2, 3)

# Create small tensors for predicted points
pred_points = torch.randn(4, 3)

# Create small tensors for ground truth normals
gt_normals = torch.randn(2, 3)

# Create small tensors for predicted normals
pred_normals = torch.randn(4, 3)

ARGS = {
    "gt_points": gt_points,
    "pred_points": pred_points,
    "gt_normals": gt_normals,
    "pred_normals": pred_normals,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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
get_points Ran
ran on a synthesized input
kacperkan/ucsgnet/ucsgnet/generators/generate_edge_data_from_point.py
code served (permissive licence) · get_code("3ebafac58ce4e2a7")
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 (0, 6) 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 torch
import numpy as np

torch.manual_seed(0)

# Create small tensors for vertices and normals
# vertices: shape (num_points, 3) - 3D coordinates
# normals: shape (num_points, 3) - normal vectors
num_points = 5
vertices = torch.randn(num_points, 3, dtype=torch.float32)
normals = torch.randn(num_points, 3, dtype=torch.float32)
# Normalize normals to unit vectors
normals = normals / (torch.norm(normals, dim=1, keepdim=True) + 1e-8)

ARGS = {
    "vertices": vertices,
    "normals": normals,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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 (0, 6) — shape (0, 6) 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
get_simple_dataset_paths_from_config Ran
ran on a synthesized input
kacperkan/ucsgnet/ucsgnet/utils.py
code served (permissive licence) · get_code("3392cd46110fb6ab")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 json
import tempfile
from pathlib import Path

# Create temporary files for the function
temp_dir = tempfile.mkdtemp()
processed_data_dir = Path(temp_dir) / "processed_data"
processed_data_dir.mkdir(exist_ok=True)

# Create some dummy data files
(processed_data_dir / "sample1.pt").touch()
(processed_data_dir / "sample2.pt").touch()
(processed_data_dir / "subdir").mkdir(exist_ok=True)
(processed_data_dir / "subdir" / "sample3.pt").touch()

# Create the split config JSON file
split_config = [
    "sample1.pt",
    "sample2.pt",
    "subdir/sample3.pt"
]

split_config_path = Path(temp_dir) / "split_config.json"
with open(split_config_path, 'w') as f:
    json.dump(split_config, f)

ARGS = {
    "processed_data_path": str(processed_data_dir),
    "split_config_path": str(split_config_path)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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
process_single_2d_image Ran
ran on a synthesized input
kacperkan/ucsgnet/ucsgnet/dataset.py
code served (permissive licence) · get_code("d717287309b3231a")
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 math
import typing as t
import cv2
import numpy as np
import torch
from PIL import Image
import torchvision.transforms as transforms_module

torch.manual_seed(0)

# Create a simple test image (BGR format, as expected by cv2.cvtColor)
# Using a 4x3 image with some non-zero pixels
image = np.zeros((4, 3, 3), dtype=np.uint8)
image[1:3, 1:2, :] = 200  # Add some white pixels in BGR

# Create a transform function that takes a PIL Image and returns a tensor
def simple_transform(pil_image):
    return torch.from_numpy(np.array(pil_image)).float() / 255

ARGS = {
    "image": image,
    "transforms": simple_transform,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, cv2, numpy, torch, torchvision — 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
quat_to_rot_matrix Ran
ran on a synthesized input
kacperkan/ucsgnet/ucsgnet/utils.py
code served (permissive licence) · get_code("0a85b5ce23d43dc8")
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 Tensor, ndim 3, dtype family float, shape (2, 3, 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 torch

torch.manual_seed(0)

# Create a valid quaternion tensor
# Shape: (batch_size=2, quaternion_dim=4)
# Quaternions are typically (w, x, y, z) or (x, y, z, w)
quaternions = torch.randn(2, 4)

# Normalize quaternions to have unit norm for valid rotation representation
quaternions = quaternions / quaternions.norm(dim=-1, keepdim=True)

ARGS = {
    "quaternions": quaternions
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 3, 3) — shape (2, 3, 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
quat_to_rot_matrix_numpy Ran
ran on a synthesized input
kacperkan/ucsgnet/ucsgnet/utils.py
code served (permissive licence) · get_code("78c7feffcb6fda0d")
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 3, dtype family float, shape (2, 3, 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 small batch of quaternions
# Shape: (2, 4) - 2 quaternions, each with 4 components (qr, qi, qj, qk)
quaternions = np.array([
    [1.0, 0.0, 0.0, 0.0],
    [0.7071, 0.7071, 0.0, 0.0]
], dtype=np.float32)

# Normalize to ensure they're valid quaternions
quaternions = quaternions / np.linalg.norm(quaternions, axis=-1, keepdims=True)

ARGS = {
    "quaternions": quaternions
}
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 3, dtype family float, shape (2, 3, 3) — shape (2, 3, 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
to_tensor Ran
ran on a synthesized input
kacperkan/ucsgnet/ucsgnet/generators/generate_edge_data_from_point.py
code served (permissive licence) · get_code("646a545e25900f47")
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 Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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
import torch

torch.manual_seed(0)

# Create a small numpy array to pass to to_tensor
data = np.random.randn(2, 4).astype(np.float32)

ARGS = {
    "data": data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
generate_single_image Not yet run
unverified (harvested, no recorded run)
kacperkan/ucsgnet/ucsgnet/generators/generate_csg_dataset.py
code served (permissive licence) · get_code("051a5c92c5f8ef95")
How it ran

no invocation recorded.

get_logger Not yet run
unverified (harvested, no recorded run)
kacperkan/ucsgnet/ucsgnet/common.py
code served (permissive licence) · get_code("3b46b02c3562dfe3")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 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 10 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

Signed distance field (SDF) is a prominent implicit representation of 3D meshes. Methods that are based on such representation achieved state-of-the-art 3D shape reconstruction quality. However, these methods struggle to reconstruct non-convex shapes. One remedy is to incorporate a constructive solid geometry framework (CSG) that represents a shape as a decomposition into primitives. It allows to embody a 3D shape of high complexity and non-convexity with a simple tree representation of Boolean operations. Nevertheless, existing approaches are supervised and require the entire CSG parse tree that is given upfront during the training process. On the contrary, we propose a model that extracts a CSG parse tree without any supervision -UCSG-NET. Our model predicts parameters of primitives and binarizes their SDF representation through differentiable indicator function. It is achieved jointly with discovering the structure of a Boolean operators tree. The model selects dynamically which operator combination over primitives leads to the reconstruction of high fidelity. We evaluate our method on 2D and 3D autoencoding tasks. We show that the predicted parse tree representation is interpretable and can be used in CAD software. 1

For agents

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

get_harvested_code_for_paper("2006.09102")
get_code_for_paper("2006.09102")
have("2006.09102")

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

curl https://syntology.ai/api/ran/2006.09102.json

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

[![Syntology run record](https://syntology.ai/api/ran/2006.09102.svg)](https://syntology.ai/paper/2006.09102)

Connect an agent — have() is free.