SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1701.06487 · 2017

Dirty Pixels: Towards End-to-End Image Processing and Perception

arXiv · PDF · Open in the Atlas

Code that ran

3 of 7 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 2 have no contract verdict; 1 failure of our instrument; 4 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/1701.06487.json · badge: /api/ran/1701.06487.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
princeton-computational-imaging/DirtyPixels canonical (classifier label)
classifier basis: census
3 ran · 4 no recorded run
FunctionStatusWhere it lives
inv_anscombe Ran
ran on a synthesized input
princeton-computational-imaging/DirtyPixels/nets/isp.py
code served (permissive licence) · get_code("074194387a3b5b47")
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 EagerTensor, ndim 4, dtype family float, shape (2, 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 tensorflow as tf
import numpy as np

# Set seed for reproducibility
tf.random.set_seed(0)

# Create small test tensors
batch_size = 2
dim_size = 4

# data: tensor used with tf.square, tf.sqrt operations
# Shape should match the operations in the function
# The function expects data to work with broadcasting against sigma_hat[:, None, None, :]
data = tf.constant(np.random.rand(batch_size, 3, 3, dim_size) * 2 + 1, dtype=tf.float32)

# sigma: 1D tensor of shape (batch_size,)
# Used as sigma_hat = sigma / alpha, then indexed with [:, None, None, :]
sigma = tf.constant(np.random.rand(batch_size) + 0.5, dtype=tf.float32)

# alpha: 1D tensor of shape (batch_size,)
# Used for division and indexing with [:, None, None, :] or [:, None, None, None]
alpha = tf.constant(np.random.rand(batch_size) + 0.5, dtype=tf.float32)

# scale: float parameter (default=1.0)
scale = 1.0

# unbiased: boolean parameter (default=False)
unbiased = False

# is_real_data: boolean parameter (default=False)
is_real_data = False

ARGS = {
    "data": data,
    "sigma": sigma,
    "alpha": alpha,
    "scale": scale,
    "unbiased": unbiased,
    "is_real_data": is_real_data,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 4, dtype family float, shape (2, 3, 3, 4) — shape (2, 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
lrelu Ran
ran on a synthesized input
princeton-computational-imaging/DirtyPixels/nets/unet.py
code served (permissive licence) · get_code("f406422099514e4c")
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 EagerTensor, 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 tensorflow as tf
import numpy as np

def lrelu(x):
    return tf.maximum(x * 0.2, x)

# Create a small TensorFlow tensor for testing
x_value = tf.constant([[-1.0, 0.5, -0.3, 2.0], [0.1, -0.8, 1.5, -0.2]], dtype=tf.float32)

ARGS = {
    'x': x_value
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, 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
wrapped_partial Ran
ran; our contract draft was wrong, not the code
princeton-computational-imaging/DirtyPixels/nets/mobilenet_isp.py
code served (permissive licence) · get_code("b6e06004f7efe660")
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 functools
import torch

def example_function(x, y, z=10):
    """A simple function to partially wrap"""
    return x + y + z

def another_function(a, b, c=5):
    """Another function for testing"""
    return a * b + c

torch.manual_seed(0)

ARGS = {
    "func": example_function,
    "args": (5, 3),
    "kwargs": {"z": 20}
}
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
anscombe Not yet run
unverified (harvested, no recorded run)
princeton-computational-imaging/DirtyPixels/nets/isp.py
code served (permissive licence) · get_code("841ffb13a2de6c74")
How it ran

no invocation recorded.

build_number_to_synset Not yet run
unverified (harvested, no recorded run)
princeton-computational-imaging/DirtyPixels/datasets/build_imagenet_data.py
code served (permissive licence) · get_code("c976f03875653135")
How it ran

no invocation recorded.

optimize_clones Not yet run
unverified (harvested, no recorded run)
princeton-computational-imaging/DirtyPixels/deployment/model_deploy.py
code served (permissive licence) · get_code("b118e995c81f2377")
How it ran

no invocation recorded.

upsample_and_concat Not yet run
unverified (harvested, no recorded run)
princeton-computational-imaging/DirtyPixels/nets/unet.py
code served (permissive licence) · get_code("99be22ebba093e63")
How it ran

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, 4 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 4 later) and 3 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
wrapped_partial 8 other papers
8 later
1703.07737 (2017-03)
In Defense of the Triplet Loss for Person Re-Identification
1710.06081 (2017-10)
Boosting Adversarial Attacks with Momentum
1808.00897 (2018-08)
BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation
and 5 more not named here
optimize_clones 4 other papers
1 dated earlier, 3 later
1608.06993 (2016-08)
Densely Connected Convolutional Networks
1804.10745 (2018-04)
Generalizing Across Domains via Cross-Gradient Training
1807.06521 (2018-07)
CBAM: Convolutional Block Attention Module
and 1 more not named here
lrelu 3 other papers
1 dated earlier, 2 later
1603.04042 (2016-03)
Deep Interactive Object Selection
1805.01934 (2018-05)
Learning to See in the Dark
1908.01311 (2019-08)
Fully Automatic Video Colorization with Self-Regularization and Diversity
upsample_and_concat 1 other paper
1 later
1805.01934 (2018-05)
Learning to See in the Dark

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

Real-world imaging systems acquire measurements that are degraded by noise, optical aberrations, and other imperfections that make image processing for human viewing and higher-level perception tasks challenging. Conventional cameras address this problem by compartmentalizing imaging from high-level task processing. As such, conventional imaging involves processing the RAW sensor measurements in a sequential pipeline of steps, such as demosaicking, denoising, deblurring, tone-mapping and compression. This pipeline is optimized to obtain a visually pleasing image. High-level processing, on the other hand, involves steps such as feature extraction, classification, tracking, and fusion. While this siloed design approach allows for efficient development, it also dictates compartmentalized performance metrics, without knowledge of the higher-level task of the camera system. For example, today's demosaicking and denoising algorithms are designed using perceptual image quality metrics but not with domain-specific tasks such as object detection in mind. We propose an end-to-end differentiable architecture that jointly performs demosaicking, denoising, deblurring, tone-mapping, and classification. The architecture learns processing pipelines whose outputs differ from those of existing ISPs optimized for perceptual quality, preserving fine detail at the cost of increased noise and artifacts. We demonstrate on captured and simulated data that our model substantially improves perception in low light and other challenging conditions, which is imperative for real-world applications. Finally, we found that the proposed model also achieves state-of-the-art accuracy when optimized for image reconstruction in low-light conditions, validating the architecture itself as a potentially useful drop-in network for reconstruction and analysis tasks beyond the applications demonstrated in this work.

For agents

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

get_harvested_code_for_paper("1701.06487")
get_code_for_paper("1701.06487")
have("1701.06487")

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

curl https://syntology.ai/api/ran/1701.06487.json

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

[![Syntology run record](https://syntology.ai/api/ran/1701.06487.svg)](https://syntology.ai/paper/1701.06487)

Connect an agent — have() is free.