SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1901.05103 · 2019

DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation

Jeong Park, Richard Newcombe, Peter Florence, Julian Straub, Steven Lovegrove

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 0 failures of our instrument; 1 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/1901.05103.json · badge: /api/ran/1901.05103.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
marian42/mesh_to_sdf community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
Facebookresearch/deepsdf community
Papers with Code archive
1 ran · 0 no recorded run
maurock/deepsdf community
Papers with Code archive
1 ran · 0 no recorded run
crazyleg/workshop-3d-neural community
Papers with Code archive
0 ran · 1 no recorded run
FunctionStatusWhere it lives
Decoder Ran
ran on a synthesized input · constructed an object; no result was computed
Facebookresearch/deepsdf/networks/deep_sdf_decoder.py
code served (permissive licence) · get_code("a897ce43d68fdf8a")
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 an instance of its own class, Decoder: the run constructed the object and did not call it, so no result was computed. 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Define the arguments dictionary
ARGS = {
    'latent_size': 4,
    'dims': [8, 16, 8],
    'dropout': [1, 2],  # list of layer indices where dropout is applied
    'dropout_prob': 0.5,
    'norm_layers': (1, 2),  # tuple of layer indices with normalization
    'latent_in': (2,),  # tuple of layer indices where latent is concatenated
    'weight_norm': False,
    'xyz_in_all': True,
    'use_tanh': True,
    'latent_dropout': True,
}
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 Decoder — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Decoder 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
SDFModel Ran
ran on a synthesized input · constructed an object; no result was computed
maurock/deepsdf/model/model_sdf.py
code served (permissive licence) · get_code("e98fabf1c2666750")
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 an instance of its own class, SDFModel: the run constructed the object and did not call it, so no result was computed. 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Construct SDFModel arguments
ARGS = {
    "num_layers": 8,
    "skip_connections": True,
    "latent_size": 4,
    "inner_dim": 64,
    "output_dim": 1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SDFModel — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a SDFModel 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
get_camera_transform Ran
ran on a synthesized input
marian42/mesh_to_sdf/mesh_to_sdf/scan.py
code served (permissive licence) · get_code("54c0331de693e44b")
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 (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 numpy as np

# Construct arguments for get_camera_transform
position = np.array([1.0, 2.0, 3.0], dtype=np.float32)
look_direction = np.array([0.0, 0.0, 1.0], dtype=np.float32)

ARGS = {
    "position": position,
    "look_direction": look_direction,
}
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 (4, 4) — shape (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
get_camera_transform_looking_at_origin Ran
ran on a synthesized input
marian42/mesh_to_sdf/mesh_to_sdf/scan.py
code served (permissive licence) · get_code("51be2c849f8505ec")
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 (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 numpy as np
from scipy.spatial.transform import Rotation

def get_rotation_matrix(angle, axis='y'):
    matrix = np.identity(4)
    if hasattr(Rotation, 'as_matrix'):
        matrix[:3, :3] = Rotation.from_euler(axis, angle).as_matrix()
    else:
        matrix[:3, :3] = Rotation.from_euler(axis, angle).as_dcm()
    return matrix

def get_camera_transform_looking_at_origin(rotation_y, rotation_x, camera_distance=2):
    camera_transform = np.identity(4)
    camera_transform[2, 3] = camera_distance
    camera_transform = np.matmul(get_rotation_matrix(rotation_x, axis='x'), camera_transform)
    camera_transform = np.matmul(get_rotation_matrix(rotation_y, axis='y'), camera_transform)
    return camera_transform

ARGS = {
    'rotation_y': 0.5,
    'rotation_x': 0.3,
    'camera_distance': 2.5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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 (4, 4) — shape (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
get_raster_points Ran
ran on a synthesized input
marian42/mesh_to_sdf/mesh_to_sdf/utils.py
code served (permissive licence) · get_code("14251929bd1ff9a9")
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 (64, 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 functools
import numpy as np

# The function takes a single parameter: voxel_resolution
# which is used as an integer argument to np.linspace
# The function creates a 3D grid of points and returns them as a float32 array

ARGS = {
    "voxel_resolution": 4
}
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 (64, 3) — shape (64, 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
get_rotation_matrix Ran
ran on a synthesized input
marian42/mesh_to_sdf/mesh_to_sdf/scan.py
code served (permissive licence) · get_code("912b298e61785af6")
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 (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 numpy as np
from scipy.spatial.transform import Rotation

ARGS = {
    'angle': 45,
    'axis': 'y'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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 (4, 4) — shape (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
SDF Not yet run
unverified (harvested, no recorded run)
crazyleg/workshop-3d-neural/final_train.py
pointer only (licence: NONE) · get_code("65f4f073c9b62ae8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, crazyleg/workshop-3d-neural when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 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 7 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

Figure 1: DeepSDF represents signed distance functions (SDFs) of shapes via latent code-conditioned feed-forward decoder networks. Above images are raycast renderings of DeepSDF interpolating between two shapes in the learned shape latent space. Best viewed digitally.

For agents

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

get_harvested_code_for_paper("1901.05103")
get_code_for_paper("1901.05103")
have("1901.05103")

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

curl https://syntology.ai/api/ran/1901.05103.json

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

[![Syntology run record](https://syntology.ai/api/ran/1901.05103.svg)](https://syntology.ai/paper/1901.05103)

Connect an agent — have() is free.