SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.12052 · 2021

Volume Rendering of Neural Implicit Surfaces

arXiv · PDF · Open in the Atlas

Code that ran

9 of 9 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 4 have no contract verdict; 5 failures of our instrument; 0 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/2106.12052.json · badge: /api/ran/2106.12052.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
lioryariv/volsdf community (classifier label)
classifier basis not recorded on the link
7 ran · 0 no recorded run
ventusff/neurecon reimplementation (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
error_bound Ran
ran; our contract draft was wrong, not the code
ventusff/neurecon/models/frameworks/volsdf.py
code served (permissive licence) · get_code("67ea58e216d0a8af")
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, 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
import numpy as np

torch.manual_seed(0)

# Create tensors with distinct shapes for different dimensions
# d_vals: [..., N_pts] - distance values at sample points
# sdf: [..., N_pts] - signed distance function values
# Following the docstring, these should have the same shape

batch_size = 2
n_pts = 4

d_vals = torch.linspace(0, 1, n_pts).unsqueeze(0).expand(batch_size, -1).float()
sdf = torch.randn(batch_size, n_pts, dtype=torch.float32)

# alpha and beta are scalar parameters for the sdf_to_sigma conversion
alpha = 1.0
beta = 0.1

ARGS = {
    "d_vals": d_vals,
    "sdf": sdf,
    "alpha": alpha,
    "beta": beta,
}
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, 3) — shape (2, 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_class Ran
ran; our contract draft was wrong, not the code
lioryariv/volsdf/code/utils/general.py
code served (permissive licence) · get_code("34bcb89c25dc2364")
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 type. 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

# The function get_class takes a string parameter 'kls' that represents
# a fully qualified class path (e.g., 'module.submodule.ClassName')
# It splits the path, imports the module, and returns the class.

# To test this, we need a valid class path string.
# We can use a built-in class from torch or a standard library.
# For example: 'torch.nn.Linear' or 'torch.optim.Adam'

ARGS = {
    'kls': 'torch.nn.Linear'
}
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 type — a type output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a type 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_embedder Ran
ran; our contract draft was wrong, not the code
lioryariv/volsdf/code/model/embedder.py
code served (permissive licence) · get_code("bb0e12df9c0d2a05")
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)

ARGS = {
    'multires': 4,
    'input_dims': 3
}
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_psnr Ran
ran on a synthesized input
lioryariv/volsdf/code/utils/rend_util.py
code served (permissive licence) · get_code("9c46582e7d919f82")
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 1, dtype family float, shape (1,) 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 torch

torch.manual_seed(0)

# Create two small image tensors
# Images should be in range [-1, 1] or [0, 1] depending on normalize_rgb
img1 = torch.randn(2, 3, 4, 4)  # batch=2, channels=3, height=4, width=4
img2 = torch.randn(2, 3, 4, 4)

ARGS = {
    'img1': img1,
    'img2': img2,
    'normalize_rgb': False
}
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 1, dtype family float, shape (1,) — shape (1,) 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
glob_imgs Ran
ran on a synthesized input
lioryariv/volsdf/code/utils/general.py
code served (permissive licence) · get_code("7bfc82ae7b496ddd")
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 os
import tempfile
from pathlib import Path

# Create a temporary directory with some test image files
temp_dir = tempfile.mkdtemp()

# Create dummy image files with various extensions
Path(os.path.join(temp_dir, "image1.png")).touch()
Path(os.path.join(temp_dir, "image2.jpg")).touch()
Path(os.path.join(temp_dir, "image3.JPEG")).touch()
Path(os.path.join(temp_dir, "image4.JPG")).touch()
Path(os.path.join(temp_dir, "other.txt")).touch()

ARGS = {
    "path": temp_dir
}
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
load_K_Rt_from_P Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
lioryariv/volsdf/code/utils/rend_util.py
pointer only (licence: MIT); the stored copy get_code serves, from felixcheng97/agap, is pointer only (licence: GPL-3.0) · get_code("d250a3f916e450fd")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from felixcheng97/agap, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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. A run is recorded.

load_rgb Ran
ran on a synthesized input
lioryariv/volsdf/code/utils/rend_util.py
code served (permissive licence) · get_code("9bc9253174730d9a")
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 (3, 3, 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 tempfile
import numpy as np
from PIL import Image
import imageio

# Create a temporary RGB image file
temp_dir = tempfile.mkdtemp()
temp_path = f"{temp_dir}/test_image.png"

# Create a simple RGB image (3x4x3 for height x width x channels)
img_array = np.random.randint(0, 256, (3, 4, 3), dtype=np.uint8)
Image.fromarray(img_array).save(temp_path)

ARGS = {
    "path": temp_path,
    "normalize_rgb": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, imageio, numpy, skimage — 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 (3, 3, 4) — shape (3, 3, 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_to_sigma Ran
ran; our contract draft was wrong, not the code
ventusff/neurecon/models/frameworks/volsdf.py
code served (permissive licence) · get_code("dd945f605d2e4471")
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. 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 small SDF tensor with arbitrary shape prefix
# Shape: (2, 4) - batch of 2, dimension 4
sdf = torch.randn(2, 4)

# Define scalar hyperparameters
alpha = 0.1
beta = 0.05

ARGS = {
    "sdf": sdf,
    "alpha": alpha,
    "beta": beta,
}
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 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
split_input Ran
ran on a synthesized input
lioryariv/volsdf/code/utils/general.py
code served (permissive licence) · get_code("c22c91faf31fd491")
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. 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 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 model_input dictionary with required keys
model_input = {
    'uv': torch.randn(2, 100, 4),  # batch=2, total_pixels=100, dim=4
    'object_mask': torch.randint(0, 2, (2, 100, 1))  # batch=2, total_pixels=100
}

# total_pixels should match the second dimension of 'uv'
total_pixels = 100

# n_pixels parameter for splitting
n_pixels = 10000

ARGS = {
    'model_input': model_input,
    'total_pixels': total_pixels,
    'n_pixels': n_pixels
}
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 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

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 5 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 1 the same month, 4 later) and 4 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.

FunctionAlso present inEarliest-datedOthers named
get_embedder 22 other papers
1 the same month, 21 later · 3 only as an assembly brick
2106.10689 (2021-06)
NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction
2205.15848 (2022-05)
Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstruction
2206.00665 (2022-06)
MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstruction
and 19 more not named here
load_K_Rt_from_P 5 other papers
5 later
2312.06657 (2023-12)
Learning Naturally Aggregated Appearance for Efficient 3D Editing
2312.12726 (2023-12)
Reducing Shape-Radiance Ambiguity in Radiance Fields with a Closed-Form Color Estimation Method
aaai_27938 (2024 (year only))
Evaluate Geometry of Radiance Fields with Low-Frequency Color Prior
and 2 more not named here
get_class 2 other papers
1 dated earlier, 1 later
2003.09852 (2020-03)
Multiview Neural Surface Reconstruction by Disentangling Geometry and Appearance
2312.04803 (2023-12)
SuperNormal: Neural Surface Reconstruction via Multi-View Normal Integration
get_psnr 1 other paper
1 later
2205.15848 (2022-05)
Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstruction
glob_imgs 1 other paper
1 dated earlier
2003.09852 (2020-03)
Multiview Neural Surface Reconstruction by Disentangling Geometry and Appearance

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

Neural volume rendering became increasingly popular recently due to its success in synthesizing novel views of a scene from a sparse set of input images. So far, the geometry learned by neural volume rendering techniques was modeled using a generic density function. Furthermore, the geometry itself was extracted using an arbitrary level set of the density function leading to a noisy, often low fidelity reconstruction. The goal of this paper is to improve geometry representation and reconstruction in neural volume rendering. We achieve that by modeling the volume density as a function of the geometry. This is in contrast to previous work modeling the geometry as a function of the volume density. In more detail, we define the volume density function as Laplace's cumulative distribution function (CDF) applied to a signed distance function (SDF) representation. This simple density representation has three benefits: (i) it provides a useful inductive bias to the geometry learned in the neural volume rendering process; (ii) it facilitates a bound on the opacity approximation error, leading to an accurate sampling of the viewing ray. Accurate sampling is important to provide a precise coupling of geometry and radiance; and (iii) it allows efficient unsupervised disentanglement of shape and appearance in volume rendering. Applying this new density representation to challenging scene multiview datasets produced high quality geometry reconstructions, outperforming relevant baselines. Furthermore, switching shape and appearance between scenes is possible due to the disentanglement of the two.

For agents

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

get_harvested_code_for_paper("2106.12052")
get_code_for_paper("2106.12052")
have("2106.12052")

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

curl https://syntology.ai/api/ran/2106.12052.json

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

[![Syntology run record](https://syntology.ai/api/ran/2106.12052.svg)](https://syntology.ai/paper/2106.12052)

Connect an agent — have() is free.