SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1904.00637 · 2019

Single Image Reflection Removal Exploiting Misaligned Training Data and Network Enhancements

arXiv · PDF · Open in the Atlas

Code that ran

5 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; 1 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 3 failures of our instrument; 6 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/1904.00637.json · badge: /api/ran/1904.00637.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/ERRNet canonical (classifier label)
classifier basis: census
5 ran · 6 no recorded run
FunctionStatusWhere it lives
compute_gradient Ran
ran; our contract draft was wrong, not the code
Vandermode/ERRNet/models/losses.py
code served (permissive licence) · get_code("91bda7c651d2e3f7")
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) 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 torch

torch.manual_seed(0)

# The function compute_gradient takes an image tensor and computes gradients
# along spatial dimensions. It expects a tensor with shape (..., height, width)
# and performs finite differences along the last two dimensions.

# Create a small image tensor with shape (batch=2, channels=3, height=4, width=5)
img = torch.randn(2, 3, 4, 5)

ARGS = {
    "img": img
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — 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
crop_quarters Ran
ran on a synthesized input
Vandermode/ERRNet/models/CX/CX_helper.py
code served (permissive licence) · get_code("7a92eb4c93e6b804")
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 (8, 2, 3, 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 tensorflow as tf
import numpy as np

# Create a small feature tensor with shape [N, fH, fW, fC]
# N=2 (batch), fH=4 (height), fW=6 (width), fC=3 (channels)
feature_tensor = tf.constant(
    np.random.randn(2, 4, 6, 3).astype(np.float32)
)

ARGS = {
    "feature_tensor": feature_tensor
}
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 (8, 2, 3, 3) — shape (8, 2, 3, 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_norm_layer Ran
ran; our contract draft was wrong, not the code
Vandermode/ERRNet/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
receptive_field Ran
ran, honoured the contract we drafted
Vandermode/ERRNet/models/networks.py
pointer only (licence: MIT); the stored copy get_code serves, from JiaxiaoWang/EvReflection, is pointer only (licence: NONE) · get_code("805fb9818f986a0d")
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 JiaxiaoWang/EvReflection, and GitHub detected no licence in that repository 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. A run is recorded.

tensor2im Ran
ran; our contract draft was wrong, not the code
Vandermode/ERRNet/models/errnet_model.py
code served (permissive licence) · get_code("fde6f58552ebe4f1")
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 (4, 3, 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 a small image tensor with shape (batch=1, channels=1, height=4, width=3)
# The function expects a tensor with batch dimension
image_tensor = torch.rand(1, 1, 4, 3, requires_grad=False)

ARGS = {
    "image_tensor": image_tensor,
    "imtype": np.uint8
}
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 (4, 3, 3) — shape (4, 3, 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
CX_loss Not yet run
unverified (harvested, no recorded run)
Vandermode/ERRNet/models/CX/CX_distance.py
code served (permissive licence) · get_code("573d768198fda403")
How it ran

no invocation recorded.

define_D Not yet run
unverified (harvested, no recorded run)
Vandermode/ERRNet/models/networks.py
code served (permissive licence) · get_code("2c137d3584064330")
How it ran

no invocation recorded.

init_loss Not yet run
unverified (harvested, no recorded run)
Vandermode/ERRNet/models/losses.py
code served (permissive licence) · get_code("310ba2102b406358")
How it ran

no invocation recorded.

random_pooling Not yet run
unverified (harvested, no recorded run)
Vandermode/ERRNet/models/CX/CX_helper.py
code served (permissive licence) · get_code("07f5bda268e05ef2")
How it ran

no invocation recorded.

random_sampling Not yet run
unverified (harvested, no recorded run)
Vandermode/ERRNet/models/CX/CX_helper.py
code served (permissive licence) · get_code("5b0d0624510bbe25")
How it ran

no invocation recorded.

symetric_CX_loss Not yet run
unverified (harvested, no recorded run)
Vandermode/ERRNet/models/CX/CX_distance.py
code served (permissive licence) · get_code("33442b837e557c99")
How it ran

no invocation recorded.

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, 6 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 5 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
2 dated earlier, 6 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
2004.05805 (2020-04)
Diversity Helps: Unsupervised Few-shot Learning via Distribution Shift-based Data Augmentation
and 5 more not named here
compute_gradient 2 other papers
2 later
2110.10546 (2021-10)
Trash or Treasure? An Interactive Dual-Stream Strategy for Single Image Reflection Separation
2111.15557 (2021-11)
Low-light Image Enhancement via Breaking Down the Darkness
define_D 2 other papers
2 later
2110.10546 (2021-10)
Trash or Treasure? An Interactive Dual-Stream Strategy for Single Image Reflection Separation
2308.10027 (2023-08)
Single Image Reflection Separation via Component Synergy
receptive_field 2 other papers
2 later
2110.10546 (2021-10)
Trash or Treasure? An Interactive Dual-Stream Strategy for Single Image Reflection Separation
2608.06184 (2026-08)
EvReflection: Event-Driven Micro-Dynamics for Reflection Removal
symetric_CX_loss 2 other papers
2 dated earlier
1803.02077 (2018-03)
The Contextual Loss for Image Transformation with Non-Aligned Data
1803.04626 (2018-03)
Maintaining Natural Image Statistics with the Contextual Loss
tensor2im 2 other papers
2 later
2110.10546 (2021-10)
Trash or Treasure? An Interactive Dual-Stream Strategy for Single Image Reflection Separation
2308.10027 (2023-08)
Single Image Reflection Separation via Component Synergy

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

Removing undesirable reflections from a single image captured through a glass window is of practical importance to visual computing systems. Although state-of-the-art methods can obtain decent results in certain situations, performance declines significantly when tackling more general real-world cases. These failures stem from the intrinsic difficulty of single image reflection removal -- the fundamental ill-posedness of the problem, and the insufficiency of densely-labeled training data needed for resolving this ambiguity within learning-based neural network pipelines. In this paper, we address these issues by exploiting targeted network enhancements and the novel use of misaligned data. For the former, we augment a baseline network architecture by embedding context encoding modules that are capable of leveraging high-level contextual clues to reduce indeterminacy within areas containing strong reflections. For the latter, we introduce an alignment-invariant loss function that facilitates exploiting misaligned real-world training data that is much easier to collect. Experimental results collectively show that our method outperforms the state-of-the-art with aligned data, and that significant improvements are possible when using additional misaligned data.

For agents

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

get_harvested_code_for_paper("1904.00637")
get_code_for_paper("1904.00637")
have("1904.00637")

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

curl https://syntology.ai/api/ran/1904.00637.json

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

[![Syntology run record](https://syntology.ai/api/ran/1904.00637.svg)](https://syntology.ai/paper/1904.00637)

Connect an agent — have() is free.