SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.11965 · 2020

CAD-Deform: Deformable Fitting of CAD Models to 3D Scans

arXiv · PDF · Open in the Atlas

Code that ran

5 of 8 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 5 have no contract verdict; 0 failures of our instrument; 3 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/2007.11965.json · badge: /api/ran/2007.11965.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
alexeybokhovkin/CAD-Deform canonical (classifier label)
classifier basis: census
5 ran · 3 no recorded run
FunctionStatusWhere it lives
edges_deformation_from_vertices Ran
ran on a synthesized input
alexeybokhovkin/CAD-Deform/shapefit/deform/src/deformation/transformation.py
code served (permissive licence) · get_code("55b95fe2d8bdceb7")
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 4, 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 testing
# torch_vertices_4d: tensor of vertices in 4D space
n_vertices = 8
torch_vertices_4d = torch.randn(n_vertices, 4, dtype=torch.float32)

# unique_edges: tensor of edge indices
n_edges = 4
unique_edges = torch.randint(0, n_vertices, (n_edges, 2), dtype=torch.long)

# target_vertices: target positions for vertices
target_vertices = torch.randn(n_vertices, 4, dtype=torch.float32)

# bitriangles_map: indices of 4 vertices forming 2 triangles around each edge
# Shape: (n_edges, 4) where each row contains 4 vertex indices
bitriangles_map = torch.tensor([
    [0, 1, 2, 3],
    [1, 2, 3, 4],
    [2, 3, 4, 5],
    [3, 4, 5, 6]
], dtype=torch.long)

ARGS = {
    'torch_vertices_4d': torch_vertices_4d,
    'unique_edges': unique_edges,
    'target_vertices': target_vertices,
    'bitriangles_map': bitriangles_map,
}
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 list, ndim 4, 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
level_merger Ran
ran on a synthesized input
alexeybokhovkin/CAD-Deform/shapefit/deform/src/deformation/preproc.py
code served (permissive licence) · get_code("fc5b4b69482fc346")
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 numpy as np
import pandas as pd

# Create a minimal data_object dict with required structure
data_object = {
    'object_id': 'obj_001',
    'parts_vertices_p2p': {
        'part_a': np.array([[0.0, 1.0, 2.0], [3.0, 4.0, 5.0]]),
        'part_b': np.array([[6.0, 7.0, 8.0]]),
        'part_c': np.array([[9.0, 10.0, 11.0], [12.0, 13.0, 14.0]]),
    }
}

# Create a partnet_map dict mapping keys to tuples of (filename, value2, value3)
partnet_map = {
    'key_0': ('part_a.obj', 'val1', 'val2'),
    'key_1': ('part_b.obj', 'val3', 'val4'),
    'key_2': ('part_c.obj', 'val5', 'val6'),
}

# Create part_levels DataFrame indexed by global_id with 'map_0k_8' column
part_levels = pd.DataFrame(
    {'map_0k_8': ['category_x', 'category_x', 'category_y']},
    index=[0, 1, 2]
)

# Create global_part_ids DataFrame with object_id and part_dir_name columns
global_part_ids = pd.DataFrame({
    'object_id': ['obj_001', 'obj_001', 'obj_001'],
    'part_dir_name': ['part_a', 'part_b', 'part_c']
})

ARGS = {
    'data_object': data_object,
    'partnet_map': partnet_map,
    'part_levels': part_levels,
    'global_part_ids': global_part_ids,
    'level': 'map_0k_8'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — 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
load_sample Ran
ran on a synthesized input
alexeybokhovkin/CAD-Deform/shapefit/utils/vox_loader.py
code served (permissive licence) · get_code("3f0876dce4d29e09")
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 Vox. 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 os
import struct
import numpy as np
import tempfile

class Vox:
    def __init__(self, dims=[0, 0, 0], res=0, grid2world=None, sdf=None, pdf=None):
        self.filename = ''
        self.dimx = dims[0]
        self.dimy = dims[1]
        self.dimz = dims[2]
        self.res = res
        self.grid2world = grid2world
        self.sdf = sdf
        self.pdf = pdf

def create_test_vox_file(filename, dimx=4, dimy=3, dimz=2):
    """Create a minimal valid .df file for testing"""
    with open(filename, 'wb') as f:
        f.write(struct.pack('I', dimx))
        f.write(struct.pack('I', dimy))
        f.write(struct.pack('I', dimz))
        f.write(struct.pack('f', 0.1))
        
        grid2world = np.eye(4, dtype=np.float32)
        f.write(grid2world.tobytes())
        
        n_elems = dimx * dimy * dimz
        sdf_data = np.ones(n_elems, dtype=np.float32) * -0.05
        f.write(sdf_data.tobytes())
        
        pdf_data = np.ones(n_elems, dtype=np.float32) * 0.1
        f.write(pdf_data.tobytes())

temp_dir = tempfile.mkdtemp()
test_file = os.path.join(temp_dir, 'test_sample.df')
create_test_vox_file(test_file)

ARGS = {
    'filename': test_file
}
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 Vox — a Vox output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Vox output the recorded ndim 0 and dtype family float 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
rts_to_matrix Ran
ran on a synthesized input
alexeybokhovkin/CAD-Deform/shapefit/utils/utils.py
code served (permissive licence) · get_code("5d49fa4bc69b429a")
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, 4, 4) 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)

# rts_to_matrix expects a tensor of shape (n, 9) where:
# - columns 0:3 are angle-axis rotations
# - columns 3:6 are translations
# - columns 6:9 are scales

batch_size = 2
rts = torch.randn(batch_size, 9, dtype=torch.double)

ARGS = {
    "rts": rts
}
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, 4, 4) — shape (2, 4, 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
vox2obj Ran
ran on a synthesized input
alexeybokhovkin/CAD-Deform/shapefit/utils/vox_loader.py
code served (permissive licence) · get_code("ff51266037ecfdc0")
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 (60, 4) 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

class VoxelGrid:
    def __init__(self):
        self.dimx = 4
        self.dimy = 3
        self.dimz = 5
        # SDF field: shape (1, dimz, dimy, dimx)
        self.sdf = np.random.rand(1, self.dimz, self.dimy, self.dimx) * 0.05
        # grid2world: 4x4 transformation matrix
        self.grid2world = np.array([
            [1.0, 0.0, 0.0, 0.5],
            [0.0, 1.0, 0.0, 0.3],
            [0.0, 0.0, 1.0, 0.2],
            [0.0, 0.0, 0.0, 1.0]
        ], dtype=np.float64)
        self.res = 0.1

vox = VoxelGrid()

ARGS = {
    "vox": vox,
    "res": None
}
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 2, dtype family float, shape (60, 4) — shape (60, 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
compute_bitriangles Not yet run
unverified (harvested, no recorded run)
alexeybokhovkin/CAD-Deform/shapefit/deform/src/deformation/preproc.py
code served (permissive licence) · get_code("70fe8c5e4d93b7c4")
How it ran

no invocation recorded.

deformation_matrix Not yet run
unverified (harvested, no recorded run)
alexeybokhovkin/CAD-Deform/shapefit/deform/src/deformation/transformation.py
code served (permissive licence) · get_code("50a8fc190e043c92")
How it ran

no invocation recorded.

find_sharp_edges Not yet run
unverified (harvested, no recorded run)
alexeybokhovkin/CAD-Deform/shapefit/deform/src/deformation/preproc.py
code served (permissive licence) · get_code("df3d643e7d4d378b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 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 8 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

Shape retrieval and alignment are a promising avenue towards turning 3D scans into lightweight CAD representations that can be used for content creation such as mobile or AR/VR gaming scenarios. Unfortunately, CAD model retrieval is limited by the availability of models in standard 3D shape collections (e.g., ShapeNet). In this work, we address this shortcoming by introducing CAD-Deform, a method which obtains more accurate CAD-to-scan fits by non-rigidly deforming retrieved CAD models. Our key contribution is a new non-rigid deformation model incorporating smooth transformations and preservation of sharp features, that simultaneously achieves very tight fits from CAD models to the 3D scan and maintains the clean, high-quality surface properties of hand-modeled CAD objects. A series of thorough experiments demonstrate that our method achieves significantly tighter scan-to-CAD fits, allowing a more accurate digital replica of the scanned real-world environment while preserving important geometric features present in synthetic CAD environments.

For agents

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

get_harvested_code_for_paper("2007.11965")
get_code_for_paper("2007.11965")
have("2007.11965")

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

curl https://syntology.ai/api/ran/2007.11965.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.11965.svg)](https://syntology.ai/paper/2007.11965)

Connect an agent — have() is free.