SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2308.13404 · 2023

Relighting Neural Radiance Fields with Shadow and Highlight Hints

arXiv · PDF · Open in the Atlas

Code that ran

11 of 11 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; 3 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 8 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/2308.13404.json · badge: /api/ran/2308.13404.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
iamNCJ/NRHints canonical (classifier label)
classifier basis: census
11 ran · 0 no recorded run
FunctionStatusWhere it lives
components_from_spherical_harmonics Ran
ran; our contract draft was wrong, not the code
iamNCJ/NRHints/fields/encodings.py
code served (permissive licence) · get_code("5d09cdd1a172883b")
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, 9) 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)

# Construct arguments for components_from_spherical_harmonics
levels = 3
directions = torch.randn(2, 4, 3)  # shape: (batch=2, seq=4, spatial=3)

ARGS = {
    "levels": levels,
    "directions": directions,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jaxtyping, 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, 9) — shape (2, 4, 9) 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_3d_gaussian Ran
ran; our contract draft was wrong, not the code
iamNCJ/NRHints/camera/ray_utils.py
code served (permissive licence) · get_code("8cbc323c7d1340e2")
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 Gaussians. 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
from dataclasses import dataclass

torch.manual_seed(0)

@dataclass
class Gaussians:
    """Stores Gaussians

    Args:
        mean: Mean of multivariate Gaussian
        cov: Covariance of multivariate Gaussian.
    """
    mean: torch.Tensor
    cov: torch.Tensor

# Create small tensors with batch size 2 and dimension 3
batch_size = 2
dim = 3

directions = torch.randn(batch_size, dim)
means = torch.randn(batch_size, dim)
dir_variance = torch.rand(batch_size, 1)
radius_variance = torch.rand(batch_size, 1)

ARGS = {
    "directions": directions,
    "means": means,
    "dir_variance": dir_variance,
    "radius_variance": radius_variance,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jaxtyping, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Gaussians — a Gaussians output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Gaussians 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
conical_frustum_to_gaussian Ran
ran; our contract draft was wrong, not the code
iamNCJ/NRHints/camera/ray_utils.py
code served (permissive licence) · get_code("9ef31786380a0afc")
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 Gaussians. 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

torch.manual_seed(0)

# Create small tensors with distinct batch and dimension sizes
batch_size = 2
spatial_dim = 3

# origins: shape (batch, 3)
origins = torch.randn(batch_size, spatial_dim)

# directions: shape (batch, 3)
directions = torch.randn(batch_size, spatial_dim)

# starts: shape (batch, 1)
starts = torch.randn(batch_size, 1)

# ends: shape (batch, 1)
ends = torch.randn(batch_size, 1) + 2.0  # ensure ends > starts

# radius: shape (batch, 1)
radius = torch.randn(batch_size, 1).abs() + 0.1  # ensure positive

ARGS = {
    "origins": origins,
    "directions": directions,
    "starts": starts,
    "ends": ends,
    "radius": radius,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jaxtyping, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Gaussians — a Gaussians output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Gaussians 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
exp_map_SE3 Ran
ran; our contract draft was wrong, not the code
iamNCJ/NRHints/camera/lie_groups.py
code served (permissive licence) · get_code("71f6c29665a5d058")
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, 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)

# Construct tangent_vector: shape [b, 6] where b=2
# The function expects se(3) tangent vectors (6-dimensional: 3 linear + 3 angular)
tangent_vector = torch.randn(2, 6, dtype=torch.float32)

ARGS = {
    "tangent_vector": tangent_vector,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jaxtyping, 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, 4) — shape (2, 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
exp_map_SO3xR3 Ran
ran; our contract draft was wrong, not the code
iamNCJ/NRHints/camera/lie_groups.py
code served (permissive licence) · get_code("14a089bd3a79c249")
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, 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 tangent vector of shape [b, 6]
# First 3 elements are translations, last 3 are so(3) tangent vectors
tangent_vector = torch.randn(2, 6)

ARGS = {
    "tangent_vector": tangent_vector,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jaxtyping, 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, 4) — shape (2, 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
expected_sin Ran
ran; our contract draft was wrong, not the code
iamNCJ/NRHints/fields/encodings.py
code served (permissive licence) · get_code("fa0affa92b7ff42b")
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 small tensors for x_means and x_vars
# x_means: mean values of a normal distribution
# x_vars: variance values of a normal distribution
# Using batch size 2 and dimension 4
x_means = torch.randn(2, 4)
x_vars = torch.abs(torch.randn(2, 4))  # variance must be non-negative

ARGS = {
    "x_means": x_means,
    "x_vars": x_vars,
}
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
extract_fields Ran
ran; our contract draft was wrong, not the code
iamNCJ/NRHints/models/neus_hint_model.py
code served (permissive licence) · get_code("b68bcf62efdab61a")
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 (128, 128, 128) 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)

# Define a simple query function that takes points and returns scalar values
def query_func(pts):
    """
    Simple query function that takes points of shape (N, 3) and returns (N,) scalar values.
    This simulates a neural network or field evaluator.
    """
    # Return sum of coordinates as a simple scalar field
    return pts.sum(dim=1)

# Define parameters
bound_min = np.array([-1.0, -1.0, -1.0])
bound_max = np.array([1.0, 1.0, 1.0])
resolution = 128

ARGS = {
    "bound_min": bound_min,
    "bound_max": bound_max,
    "resolution": resolution,
    "query_func": query_func,
}
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 3, dtype family float, shape (128, 128, 128) — shape (128, 128, 128) 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
rot_phi Ran
ran, honoured the contract we drafted
iamNCJ/NRHints/camera/video_pose_utils.py
code served (permissive licence) · get_code("6f35e5c391898ce7")
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

def rot_phi(phi):
    return np.array([[1, 0, 0, 0], [0, np.cos(phi), -np.sin(phi), 0], [0, np.sin(phi), np.cos(phi), 0], [0, 0, 0, 1]], dtype=np.float32)

ARGS = {
    "phi": 0.5
}
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
rot_theta Ran
ran, honoured the contract we drafted
iamNCJ/NRHints/camera/video_pose_utils.py
code served (permissive licence) · get_code("7f9c426ffc6ab29e")
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

# The function rot_theta takes a single parameter 'th' (theta, an angle)
# and returns a 4x4 rotation matrix around the Y-axis

# Create a small angle value for testing
theta_value = np.pi / 4  # 45 degrees

ARGS = {
    "th": theta_value
}
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
sample_pdf Ran
ran; our contract draft was wrong, not the code
iamNCJ/NRHints/models/neus_hint_model.py
code served (permissive licence) · get_code("2ee5137eb749ef9f")
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 small test tensors
# bins: shape (batch_size, num_bins) - the bin edges
bins = torch.linspace(0, 1, steps=5, dtype=torch.float32).unsqueeze(0).expand(2, -1)

# weights: shape (batch_size, num_bins-1) - weights for each bin
weights = torch.ones(2, 4, dtype=torch.float32)

# N_samples: number of samples to draw
N_samples = 3

# det: deterministic sampling flag
det = False

# pytest: use numpy random instead of torch random
pytest = False

ARGS = {
    "bins": bins,
    "weights": weights,
    "N_samples": N_samples,
    "det": det,
    "pytest": pytest,
}
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
trans_t Ran
ran, honoured the contract we drafted
iamNCJ/NRHints/camera/video_pose_utils.py
code served (permissive licence) · get_code("c5433e0050d3b8e3")
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

def trans_t(t):
    return np.array([[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, t], [0, 0, 0, 1]], dtype=np.float32)

# The function takes a single parameter 't' which is used in a matrix translation
# The parameter should be a scalar value that gets placed in the [2, 3] position
# of the 4x4 transformation matrix

ARGS = {
    "t": 5.0
}
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

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 1 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.

FunctionAlso present inEarliest-datedOthers named
extract_fields 1 other paper
1 same year, order not determinable
Ling_ShadowNeuS_Neural_SDF_Reconstruction_by_Shadow_Ray_Supervision_CVPR_2023_paper (2023 (year only))
ShadowNeuS: Neural SDF Reconstruction by Shadow Ray Supervision

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

This paper presents a novel neural implicit radiance representation for free viewpoint relighting from a small set of unstructured photographs of an object lit by a moving point light source different from the view position. We express the shape as a signed distance function modeled by a multi layer perceptron. In contrast to prior relightable implicit neural representations, we do not disentangle the different reflectance components, but model both the local and global reflectance at each point by a second multi layer perceptron that, in addition, to density features, the current position, the normal (from the signed distace function), view direction, and light position, also takes shadow and highlight hints to aid the network in modeling the corresponding high frequency light transport effects. These hints are provided as a suggestion, and we leave it up to the network to decide how to incorporate these in the final relit result. We demonstrate and validate our neural implicit representation on synthetic and real scenes exhibiting a wide variety of shapes, material properties, and global illumination light transport.

For agents

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

get_harvested_code_for_paper("2308.13404")
get_code_for_paper("2308.13404")
have("2308.13404")

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

curl https://syntology.ai/api/ran/2308.13404.json

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

[![Syntology run record](https://syntology.ai/api/ran/2308.13404.svg)](https://syntology.ai/paper/2308.13404)

Connect an agent — have() is free.