SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2004.10566 · 2020

Efficient Neighbourhood Consensus Networks via Submanifold Sparse Convolutions

arXiv · PDF · Open in the Atlas

Code that ran

11 of 18 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 7 have no contract verdict; 3 failures of our instrument; 7 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/2004.10566.json · badge: /api/ran/2004.10566.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
ignacio-rocco/sparse-ncnet community (classifier label)
classifier basis not recorded on the link
11 ran · 7 no recorded run
FunctionStatusWhere it lives
MutualMatching Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ignacio-rocco/sparse-ncnet/lib/model.py
code served (permissive licence) · get_code("6143bd380b5003fc")
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 Tensor, ndim 6, dtype family float, shape (2, 1, 3, 4, 5, 6) 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 a 6D tensor with shape (batch_size, ch, fs1, fs2, fs3, fs4)
# Let's create a small tensor with distinct dimensions for clarity
batch_size = 2
ch = 1
fs1 = 3
fs2 = 4
fs3 = 5
fs4 = 6

corr4d = torch.randn(batch_size, ch, fs1, fs2, fs3, fs4)

ARGS = {
    "corr4d": corr4d
}
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 Tensor, ndim 6, dtype family float, shape (2, 1, 3, 4, 5, 6) — shape (2, 1, 3, 4, 5, 6) 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
corr_dense Ran
ran on a synthesized input
ignacio-rocco/sparse-ncnet/lib/model.py
code served (permissive licence) · get_code("53897423d99a3ceb")
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 6, dtype family float, shape (2, 1, 3, 5, 7, 6) 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 two feature tensors with distinct dimensions
# Shape: (batch, channels, height, width)
feature_A = torch.randn(2, 4, 3, 5)  # batch=2, channels=4, height=3, width=5
feature_B = torch.randn(2, 4, 7, 6)  # batch=2, channels=4, height=7, width=6

ARGS = {
    "feature_A": feature_A,
    "feature_B": feature_B,
}
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 6, dtype family float, shape (2, 1, 3, 5, 7, 6) — shape (2, 1, 3, 5, 7, 6) 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
default_collate Ran
ran on a synthesized input
ignacio-rocco/sparse-ncnet/lib/dataloader.py
code served (permissive licence) · get_code("eb5ca9515b0fe778")
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
import collections
import numpy as np

torch.manual_seed(0)

# The function expects a 'batch' parameter which is a list of items
# Let's create different test cases for the various branches

# Test case 1: batch of tensors
batch_of_tensors = [torch.randn(3, 4) for _ in range(2)]

# Test case 2: batch of numpy arrays
batch_of_numpy = [np.random.randn(3, 4).astype(np.float32) for _ in range(2)]

# Test case 3: batch of integers
batch_of_ints = [1, 2]

# Test case 4: batch of floats
batch_of_floats = [1.5, 2.5]

# Test case 5: batch of dicts
batch_of_dicts = [
    {'a': torch.randn(3, 4), 'b': torch.randn(2, 5)},
    {'a': torch.randn(3, 4), 'b': torch.randn(2, 5)}
]

# Test case 6: batch of sequences/tuples
batch_of_sequences = [
    (torch.randn(3, 4), torch.randn(2, 5)),
    (torch.randn(3, 4), torch.randn(2, 5))
]

# Pick one for the main test - using tensors as primary case
ARGS = {
    'batch': batch_of_tensors
}
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 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
featureL2Norm Ran
ran, honoured the contract we drafted
ignacio-rocco/sparse-ncnet/lib/model.py
code served (permissive licence) · get_code("cf9656ffec66503c")
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 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 feature tensor with shape (batch_size=2, feature_dim=4)
# The function expects a 2D tensor where dimension 1 is summed over
feature = torch.randn(2, 4)

ARGS = {
    "feature": feature
}
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 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
imshow_image Ran
ran on a synthesized input
ignacio-rocco/sparse-ncnet/lib/plot.py
code served (permissive licence) · get_code("8eb839e468634ab2")
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 int, shape (4, 5, 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 numpy as np

# Create a small test image with shape (3, 4, 5) - 3 channels, 4x5 spatial dims
# Using float values in a reasonable range for preprocessing
np.random.seed(0)
image = np.random.randn(3, 4, 5).astype(np.float32) * 50 + 100

ARGS = {
    'image': image,
    'preprocessing': 'torch'
}
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 3, dtype family int, shape (4, 5, 3) — shape (4, 5, 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
normalize_axis Ran
ran on a synthesized input
ignacio-rocco/sparse-ncnet/lib/point_tnf.py
code served (permissive licence) · get_code("2fa12cb30a255583")
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 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
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# def normalize_axis(x, L):
#     return (x - 1 - (L - 1) / 2) * 2 / (L - 1)
#
# Parameters:
# - x: a numeric value (scalar, array, or tensor) that will be subtracted from 1 and used in arithmetic
# - L: a numeric value representing a length, used in division and arithmetic operations

# Create valid arguments:
x = torch.tensor(5.0)  # scalar tensor value
L = 10  # integer representing length

ARGS = {
    "x": x,
    "L": L,
}
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 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
normalize_caffe Ran
ran on a synthesized input
ignacio-rocco/sparse-ncnet/lib/normalization.py
code served (permissive licence) · get_code("8793923899a03f2d")
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 4, dtype family float, shape (2, 3, 4, 5) 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 normalize_caffe takes an image tensor
# It handles both 3D and 4D tensors
# For 4D tensors: (batch, channels, height, width)
# It performs channel reordering from RGB to BGR and applies mean subtraction

# Create a sample 4D image tensor (batch=2, channels=3, height=4, width=5)
image = torch.randn(2, 3, 4, 5)

ARGS = {
    "image": image
}
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 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 4, 5) 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
normalize_image Ran
ran on a synthesized input
ignacio-rocco/sparse-ncnet/lib/normalization.py
code served (permissive licence) · get_code("68162679b97bdf55")
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 4, dtype family float, shape (2, 3, 4, 5) 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
from torch.autograd import Variable

torch.manual_seed(0)

# Test case 1: 3D tensor (C, H, W) with forward normalization
image_3d = torch.randn(3, 4, 5)

# Test case 2: 4D tensor (B, C, H, W) with forward normalization
image_4d = torch.randn(2, 3, 4, 5)

# Use the 4D case as the primary test
image = image_4d
forward = True
mean = [0.485, 0.456, 0.406]
std = [0.229, 0.224, 0.225]

ARGS = {
    "image": image,
    "forward": forward,
    "mean": mean,
    "std": std,
}
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 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 4, 5) 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
pck Ran
ran; our contract draft was wrong, not the code
ignacio-rocco/sparse-ncnet/lib/eval_util.py
code served (permissive licence) · get_code("8ea2a77a19d38458")
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 (2,) 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 torch.nn

torch.manual_seed(0)

# Analyze the function:
# - source_points: indexed as [i, :], then [0, :] and [1, :] -> 2D tensor, shape (batch_size, 2, num_points)
# - warped_points: same indexing pattern as source_points -> shape (batch_size, 2, num_points)
# - L_pck: indexed as [i] and .expand_as(point_distance) -> 1D tensor, shape (batch_size,)
# - The function checks for -1 values in source_points to determine valid points
# - point_distance has shape (num_points,) after the sum and sqrt operations

batch_size = 2
num_points = 4

# Create source_points with shape (batch_size, 2, num_points)
# Include some -1 values to test the valid point detection
source_points = torch.tensor([
    [[1.0, 2.0, 3.0, -1.0],
     [4.0, 5.0, 6.0, -1.0]],
    [[0.5, 1.5, 2.5, 3.5],
     [2.0, 3.0, 4.0, 5.0]]
], dtype=torch.float32)

# Create warped_points with similar shape
warped_points = torch.tensor([
    [[1.1, 2.1, 3.1, -1.0],
     [4.1, 5.1, 6.1, -1.0]],
    [[0.6, 1.6, 2.6, 3.6],
     [2.1, 3.1, 4.1, 5.1]]
], dtype=torch.float32)

# Create L_pck with shape (batch_size,)
# These represent reference lengths for each sample
L_pck = torch.tensor([5.0, 6.0], dtype=torch.float32)

ARGS = {
    "source_points": source_points,
    "warped_points": warped_points,
    "L_pck": L_pck,
    "alpha": 0.1
}
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 (2,) — shape (2,) 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
swig_ptr_from_FloatTensor Ran
ran; our contract draft was wrong, not the code
ignacio-rocco/sparse-ncnet/lib/knn.py
code served (permissive licence) · get_code("f295b8050ce2a4f1")
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 SwigPyObject. 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 faiss

torch.manual_seed(0)

# Create a small contiguous float32 tensor
x = torch.randn(2, 4, dtype=torch.float32).contiguous()

ARGS = {
    'x': x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
faiss, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SwigPyObject — a SwigPyObject output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SwigPyObject 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
unnormalize_axis Ran
ran on a synthesized input
ignacio-rocco/sparse-ncnet/lib/point_tnf.py
code served (permissive licence) · get_code("f78be05f4ea43f88")
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 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
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# def unnormalize_axis(x, L):
#     return x * (L - 1) / 2 + 1 + (L - 1) / 2
# 
# Parameters:
# - x: appears to be a numeric value or tensor that gets multiplied by (L - 1) / 2
# - L: appears to be a scalar representing a length/size
#
# The function performs an unnormalization operation on axis values.
# x can be a tensor or scalar
# L should be a scalar integer (axis length)

x = torch.tensor(0.5)  # A normalized value between -1 and 1
L = 5  # A dimension size

ARGS = {
    "x": x,
    "L": L
}
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 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
conv4d Not yet run
unverified (harvested, no recorded run)
ignacio-rocco/sparse-ncnet/lib/conv4d.py
code served (permissive licence) · get_code("3bb2c78a6431d327")
How it ran

no invocation recorded.

corr_to_matches Not yet run
unverified (harvested, no recorded run)
ignacio-rocco/sparse-ncnet/lib/point_tnf.py
code served (permissive licence) · get_code("33e385b9ba0dd3bf")
How it ran

no invocation recorded.

me_to_torch Not yet run
unverified (harvested, no recorded run)
ignacio-rocco/sparse-ncnet/lib/sparse.py
code served (permissive licence) · get_code("15c590dac3984ce1")
How it ran

no invocation recorded.

pin_memory_batch Not yet run
unverified (harvested, no recorded run)
ignacio-rocco/sparse-ncnet/lib/dataloader.py
code served (permissive licence) · get_code("6d8c67dc116f1f8a")
How it ran

no invocation recorded.

plot_image Not yet run
unverified (harvested, no recorded run)
ignacio-rocco/sparse-ncnet/lib/plot.py
code served (permissive licence) · get_code("632470f16cec485c")
How it ran

no invocation recorded.

search_index_pytorch Not yet run
unverified (harvested, no recorded run)
ignacio-rocco/sparse-ncnet/lib/knn.py
code served (permissive licence) · get_code("78ca4879f13cc9d5")
How it ran

no invocation recorded.

swig_ptr_from_LongTensor Not yet run
unverified (harvested, no recorded run)
ignacio-rocco/sparse-ncnet/lib/knn.py
code served (permissive licence) · get_code("892aeccb03c9a237")
How it ran

no invocation recorded.

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 18 distinct functions with an identity, 14 are also present in other papers whose code we hold (13 with at least one paper dated earlier, 0 the same month, 6 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
featureL2Norm 7 other papers
4 dated earlier, 3 later · 2 only as an assembly brick
1703.05593 (2017-03)
Convolutional neural network architecture for geometric matching
1810.10510 (2018-10)
Neighbourhood Consensus Networks
1909.11895 (2019-09)
Joint-task Self-supervised Learning for Temporal Correspondence
and 4 more not named here
MutualMatching 4 other papers
2 dated earlier, 2 later · 1 only as an assembly brick
1810.10510 (2018-10)
Neighbourhood Consensus Networks
2003.12059 (2020-03)
Correspondence Networks with Adaptive Neighbourhood Consensus
2006.08844 (2020-06)
Dual-Resolution Correspondence Networks
and 1 more not named here
default_collate 4 other papers
4 dated earlier
1703.05593 (2017-03)
Convolutional neural network architecture for geometric matching
1803.09877 (2018-03)
DRACO: Byzantine-resilient Distributed Training via Redundant Gradients
1810.10510 (2018-10)
Neighbourhood Consensus Networks
and 1 more not named here
pin_memory_batch 4 other papers
4 dated earlier
1703.05593 (2017-03)
Convolutional neural network architecture for geometric matching
1803.09877 (2018-03)
DRACO: Byzantine-resilient Distributed Training via Redundant Gradients
1810.10510 (2018-10)
Neighbourhood Consensus Networks
and 1 more not named here
swig_ptr_from_FloatTensor 4 other papers
2 dated earlier, 2 later
1805.09501 (2018-05)
AutoAugment: Learning Augmentation Policies from Data
1911.02116 (2019-11)
Unsupervised Cross-lingual Representation Learning at Scale
aaai_19970 (2022 (year only))
SECRET: Self-Consistent Pseudo Label Refinement for Unsupervised Domain Adaptive Person Re-identification
and 1 more not named here
swig_ptr_from_LongTensor 4 other papers
2 dated earlier, 2 later
1805.09501 (2018-05)
AutoAugment: Learning Augmentation Policies from Data
1911.02116 (2019-11)
Unsupervised Cross-lingual Representation Learning at Scale
aaai_19970 (2022 (year only))
SECRET: Self-Consistent Pseudo Label Refinement for Unsupervised Domain Adaptive Person Re-identification
and 1 more not named here
normalize_axis 3 other papers
3 dated earlier
1703.05593 (2017-03)
Convolutional neural network architecture for geometric matching
1810.10510 (2018-10)
Neighbourhood Consensus Networks
2003.12059 (2020-03)
Correspondence Networks with Adaptive Neighbourhood Consensus
pck 3 other papers
3 dated earlier
1703.05593 (2017-03)
Convolutional neural network architecture for geometric matching
1810.10510 (2018-10)
Neighbourhood Consensus Networks
2003.12059 (2020-03)
Correspondence Networks with Adaptive Neighbourhood Consensus
search_index_pytorch 3 other papers
1 dated earlier, 2 later
1805.09501 (2018-05)
AutoAugment: Learning Augmentation Policies from Data
aaai_19970 (2022 (year only))
SECRET: Self-Consistent Pseudo Label Refinement for Unsupervised Domain Adaptive Person Re-identification
2204.09826 (2022-04)
SimMC: Simple Masked Contrastive Learning of Skeleton Representations for Unsupervised Person Re-Identification
unnormalize_axis 3 other papers
3 dated earlier
1703.05593 (2017-03)
Convolutional neural network architecture for geometric matching
1810.10510 (2018-10)
Neighbourhood Consensus Networks
2003.12059 (2020-03)
Correspondence Networks with Adaptive Neighbourhood Consensus
corr_to_matches 2 other papers
2 dated earlier
1810.10510 (2018-10)
Neighbourhood Consensus Networks
2003.12059 (2020-03)
Correspondence Networks with Adaptive Neighbourhood Consensus
normalize_image 2 other papers
2 dated earlier
1810.10510 (2018-10)
Neighbourhood Consensus Networks
2003.12059 (2020-03)
Correspondence Networks with Adaptive Neighbourhood Consensus
plot_image 2 other papers
2 dated earlier
1810.10510 (2018-10)
Neighbourhood Consensus Networks
2003.12059 (2020-03)
Correspondence Networks with Adaptive Neighbourhood Consensus
imshow_image 1 other paper
1 later
2203.07003 (2022-03)
MTLDesc: Looking Wider to Describe Better

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

In this work we target the problem of estimating accurately localised correspondences between a pair of images. We adopt the recent Neighbourhood Consensus Networks that have demonstrated promising performance for difficult correspondence problems and propose modifications to overcome their main limitations: large memory consumption, large inference time and poorly localised correspondences. Our proposed modifications can reduce the memory footprint and execution time more than $10\times$, with equivalent results. This is achieved by sparsifying the correlation tensor containing tentative matches, and its subsequent processing with a 4D CNN using submanifold sparse convolutions. Localisation accuracy is significantly improved by processing the input images in higher resolution, which is possible due to the reduced memory footprint, and by a novel two-stage correspondence relocalisation module. The proposed Sparse-NCNet method obtains state-of-the-art results on the HPatches Sequences and InLoc visual localisation benchmarks, and competitive results in the Aachen Day-Night benchmark.

For agents

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

get_harvested_code_for_paper("2004.10566")
get_code_for_paper("2004.10566")
have("2004.10566")

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

curl https://syntology.ai/api/ran/2004.10566.json

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

[![Syntology run record](https://syntology.ai/api/ran/2004.10566.svg)](https://syntology.ai/paper/2004.10566)

Connect an agent — have() is free.