SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2108.02158 · 2021

Physics-based Noise Modeling for Extreme Low-light Photography

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 3 have no contract verdict; 2 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/2108.02158.json · badge: /api/ran/2108.02158.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
Vandermode/NoiseModel canonical (classifier label)
classifier basis: census
4 ran · 2 no recorded run
Vandermode/ELD community (classifier label)
classifier basis: plan_row
1 ran · 1 no recorded run
FunctionStatusWhere it lives
compute_expo_ratio Ran
ran on a synthesized input
Vandermode/NoiseModel/dataset/sid_dataset.py
code served (permissive licence) · get_code("e6313ae77bf51222")
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 float, ndim 0, dtype family float, shape () 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

# The function compute_expo_ratio takes two filename strings as input.
# It extracts exposure values from the filenames by:
# 1. Splitting on '_' and taking the last element
# 2. Removing the last 5 characters (file extension like ".exif")
# 3. Converting to float
# It then computes a ratio of gt_exposure / in_exposure, capped at 300.

# We need to construct valid filename strings that contain exposure values.
# Format: anything_<exposure_value>.exif

input_fn = "image_input_2.5.exif"
target_fn = "image_target_10.0.exif"

ARGS = {
    "input_fn": input_fn,
    "target_fn": target_fn,
}
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 float, ndim 0, dtype family float, shape () — shape () 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
crop_center Ran
ran on a synthesized input
Vandermode/NoiseModel/dataset/sid_dataset.py
code served (permissive licence) · get_code("103b27cc220783e8")
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 (3, 6, 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)

# The function expects:
# - img: a tensor with shape (channels, y, x)
# - cropx: integer for crop width
# - cropy: integer for crop height

# Create a small test image tensor with shape (channels, height, width)
img = torch.randn(3, 8, 12)  # 3 channels, 8 height, 12 width

# Define crop dimensions
cropx = 4
cropy = 6

ARGS = {
    "img": img,
    "cropx": cropx,
    "cropy": cropy,
}
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 (3, 6, 4) — shape (3, 6, 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_norm_layer Ran
ran; our contract draft was wrong, not the code
Vandermode/NoiseModel/models/networks.py
code served (permissive licence) · get_code("7702313c16b1182f")
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 partial. 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
import functools

torch.manual_seed(0)

# The function get_norm_layer takes one parameter: norm_type
# It returns a partial function or None based on the norm_type string
# We need to test it with valid norm_type values

ARGS = {
    'norm_type': 'instance'
}
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 partial — a partial output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a partial 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
read_dorf Ran
ran; our contract draft was wrong, not the code
Vandermode/ELD/EMoR/EMoR.py
code served (permissive licence) · get_code("031ab536c74127bb")
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 tempfile
import os
import numpy as np

# Create a temporary file with valid DORF format
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.dorf")

# Write DORF format data: curve_names on line 0, Es on line 3, Bs on line 5, repeated every 6 lines
with open(temp_file, 'w') as f:
    # First curve (lines 0-5)
    f.write("curve_1\n")
    f.write("dummy_line_1\n")
    f.write("dummy_line_2\n")
    f.write("1.0 2.0 3.0 4.0\n")
    f.write("dummy_line_4\n")
    f.write("5.0 6.0 7.0 8.0\n")
    
    # Second curve (lines 6-11)
    f.write("curve_2\n")
    f.write("dummy_line_1\n")
    f.write("dummy_line_2\n")
    f.write("9.0 10.0 11.0 12.0\n")
    f.write("dummy_line_4\n")
    f.write("13.0 14.0 15.0 16.0\n")

ARGS = {
    "address": temp_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 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
tensor2im Ran
ran on a synthesized input
Vandermode/NoiseModel/models/ELD_model.py
code served (permissive licence) · get_code("ee0657d1b9a5b876")
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, 5, 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
import numpy as np

torch.manual_seed(0)

# Case 1: Non-video mode (visualize=False, video=False)
# image_tensor needs shape [batch, channels, height, width]
# After [0] indexing: [channels, height, width]
# After transpose (1,2,0): [height, width, channels]
image_tensor_non_video = torch.randn(2, 4, 3, 5)

# Case 2: Video mode (video=True)
# image_tensor needs shape [batch, channels, height, width] for video processing
# After transpose (0,2,3,1): [batch, height, width, channels]
image_tensor_video = torch.randn(2, 4, 3, 5)

ARGS = {
    "image_tensor": image_tensor_non_video,
    "visualize": False,
    "video": False,
}
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 (3, 5, 4) — shape (3, 5, 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
init_loss Not yet run
unverified (harvested, no recorded run)
Vandermode/NoiseModel/models/losses.py
code served (permissive licence) · get_code("960d32e3ed442108")
How it ran

no invocation recorded.

read_emor Not yet run
unverified (harvested, no recorded run)
Vandermode/ELD/EMoR/EMoR.py
code served (permissive licence) · get_code("dfa0c71130d11161")
How it ran

no invocation recorded.

receptive_field Not yet run
unverified (harvested, no recorded run)
Vandermode/NoiseModel/models/networks.py
code served (permissive licence) · get_code("683ebeade7fc133f")
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, 3 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 3 later) and 5 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_norm_layer 8 other papers
5 dated earlier, 3 later
1804.03999 (2018-04)
Attention U-Net: Learning Where to Look for the Pancreas
1903.08682 (2019-03)
Im2Pencil: Controllable Pencil Illustration from Photographs
1904.00637 (2019-04)
Single Image Reflection Removal Exploiting Misaligned Training Data and Network Enhancements
and 5 more not named here
read_dorf 1 other paper
1 later
2307.16508 (2023-07)
Towards General Low-Light Raw Noise Synthesis and Modeling
read_emor 1 other paper
1 later
2307.16508 (2023-07)
Towards General Low-Light Raw Noise Synthesis and Modeling

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

Enhancing the visibility in extreme low-light environments is a challenging task. Under nearly lightless condition, existing image denoising methods could easily break down due to significantly low SNR. In this paper, we systematically study the noise statistics in the imaging pipeline of CMOS photosensors, and formulate a comprehensive noise model that can accurately characterize the real noise structures. Our novel model considers the noise sources caused by digital camera electronics which are largely overlooked by existing methods yet have significant influence on raw measurement in the dark. It provides a way to decouple the intricate noise structure into different statistical distributions with physical interpretations. Moreover, our noise model can be used to synthesize realistic training data for learning-based low-light denoising algorithms. In this regard, although promising results have been shown recently with deep convolutional neural networks, the success heavily depends on abundant noisy clean image pairs for training, which are tremendously difficult to obtain in practice. Generalizing their trained models to images from new devices is also problematic. Extensive experiments on multiple low-light denoising datasets -- including a newly collected one in this work covering various devices -- show that a deep neural network trained with our proposed noise formation model can reach surprisingly-high accuracy. The results are on par with or sometimes even outperform training with paired real data, opening a new door to real-world extreme low-light photography.

For agents

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

get_harvested_code_for_paper("2108.02158")
get_code_for_paper("2108.02158")
have("2108.02158")

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

curl https://syntology.ai/api/ran/2108.02158.json

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

[![Syntology run record](https://syntology.ai/api/ran/2108.02158.svg)](https://syntology.ai/paper/2108.02158)

Connect an agent — have() is free.