SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2005.01014 · 2020

Feature-metric Registration: A Fast Semi-supervised Approach for Robust Point Cloud Registration without Correspondences

arXiv · PDF · Open in the Atlas

Code that ran

14 of 17 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 14 have no contract verdict; 0 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/2005.01014.json · badge: /api/ran/2005.01014.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
XiaoshuiHuang/fmr canonical (classifier label)
classifier basis: census
14 ran · 3 no recorded run
FunctionStatusWhere it lives
batch_inverse Ran
ran on a synthesized input
XiaoshuiHuang/fmr/se_math/invmat.py
code served (permissive licence) · get_code("62e0a53461028a1f")
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, 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 batch of invertible square matrices
# batch_size=2, h=w=4 (small square matrices)
batch_size = 2
matrix_size = 4

# Create random matrices and ensure they are invertible
x = torch.randn(batch_size, matrix_size, matrix_size)

# Make them more likely to be invertible by adding to diagonal
for i in range(batch_size):
    x[i] = x[i] + torch.eye(matrix_size) * 5

ARGS = {
    "x": x
}
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 (2, 4, 4) — shape (2, 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
batch_inverse_dx Ran
ran on a synthesized input
XiaoshuiHuang/fmr/se_math/invmat.py
code served (permissive licence) · get_code("4fbdd532c7b1a6d6")
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 5, dtype family float, shape (2, 4, 4, 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 torch
import torch.nn as nn

torch.manual_seed(0)

# The function batch_inverse_dx takes a tensor y and performs operations on it.
# It expects y to have shape (batch_size, h, w) where h == w
# Let's create a small valid tensor with batch_size=2, h=w=4

batch_size = 2
h = 4
w = 4

y = torch.randn(batch_size, h, w)

ARGS = {
    "y": y
}
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 5, dtype family float, shape (2, 4, 4, 4, 4) — shape (2, 4, 4, 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
cross_prod Ran
ran on a synthesized input
XiaoshuiHuang/fmr/se_math/so3.py
code served (permissive licence) · get_code("bb80e2d7df517e02")
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

torch.manual_seed(0)

# The function cross_prod takes two parameters: x and y
# It reshapes them to (-1, 3), computes cross product along dim=1, then reshapes back
# So x and y must have a total number of elements divisible by 3
# Let's create two tensors with shape (2, 3) to match the cross product requirement

x = torch.randn(2, 3)
y = torch.randn(2, 3)

ARGS = {
    "x": x,
    "y": y,
}
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, 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
flatten Ran
ran on a synthesized input
XiaoshuiHuang/fmr/model.py
code served (permissive licence) · get_code("aaa5e6d5223ca869")
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, 60) 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 flatten(x) calls x.view(x.size(0), -1)
# So x must be a tensor with a .view() method and .size() method
# We need a multi-dimensional tensor to make flattening meaningful

x = torch.randn(2, 4, 3, 5)  # batch_size=2, then 3 other dimensions

ARGS = {
    "x": x
}
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, 60) — shape (2, 60) 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
liebracket Ran
ran on a synthesized input
XiaoshuiHuang/fmr/se_math/so3.py
code served (permissive licence) · get_code("a253a6362a8213c9")
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

torch.manual_seed(0)

# liebracket takes two parameters: x and y
# The function calls cross_prod on both, which:
# 1. Reshapes to (-1, 3)
# 2. Calls torch.cross with dim=1
# 3. Reshapes back to original shape
#
# This means x and y must have shapes compatible with reshaping to (-1, 3)
# and back. A batch of 2D vectors with 3 components works: shape (2, 3)

x = torch.randn(2, 3)
y = torch.randn(2, 3)

ARGS = {
    "x": x,
    "y": y
}
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, 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
mat Ran
ran on a synthesized input
XiaoshuiHuang/fmr/se_math/se3.py
code served (permissive licence) · get_code("92b15d1f75b10f1b")
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, 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 a tensor x with shape (..., 6)
# where the last dimension has 6 elements: w1, w2, w3, v1, v2, v3
# The function reshapes it to (-1, 6), processes, then reshapes back to (..., 4, 4)

# Create a small tensor with batch size 2 and last dimension 6
x = torch.randn(2, 6)

ARGS = {
    "x": x
}
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 (2, 4, 4) — shape (2, 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
mat Ran
ran on a synthesized input
XiaoshuiHuang/fmr/se_math/so3.py
code served (permissive licence) · get_code("9d3e60944e1061f1")
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, 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

torch.manual_seed(0)

# The function `mat` takes a tensor x, reshapes it to (-1, 3),
# extracts 3 columns, and constructs a skew-symmetric matrix.
# The input should have last dimension 3, and we can add batch dimensions.
# Let's create a tensor with shape (2, 4, 3) - batch of 2, 4 vectors, 3 components each

x = torch.randn(2, 4, 3)

ARGS = {
    'x': x
}
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, 4, 3, 3) — shape (2, 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
objread Ran
ran on a synthesized input
XiaoshuiHuang/fmr/se_math/mesh.py
code served (permissive licence) · get_code("52a9f7d06b228b77")
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 Mesh. 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

# Create a temporary OBJ file for testing
temp_dir = tempfile.mkdtemp()
obj_filepath = os.path.join(temp_dir, "test.obj")

# Write a simple OBJ file with vertices and faces
obj_content = """# Simple test cube
v 0.0 0.0 0.0
v 1.0 0.0 0.0
v 1.0 1.0 0.0
v 0.0 1.0 0.0
v 0.0 0.0 1.0
v 1.0 0.0 1.0
v 1.0 1.0 1.0
v 0.0 1.0 1.0
f 1 2 3
f 1 3 4
f 5 6 7
f 5 7 8
"""

with open(obj_filepath, 'w') as f:
    f.write(obj_content)

ARGS = {
    'filepath': obj_filepath,
    'points_only': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, mpl_toolkits, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Mesh — a Mesh output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Mesh 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
offread Ran
ran on a synthesized input
XiaoshuiHuang/fmr/se_math/mesh.py
code served (permissive licence) · get_code("e3a74959859b332c")
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 Mesh. 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

# Create a temporary OFF file for testing
temp_dir = tempfile.mkdtemp()
off_filepath = os.path.join(temp_dir, "test_mesh.off")

# Write a simple valid OFF file
with open(off_filepath, 'w') as f:
    f.write("OFF\n")
    f.write("4 1 0\n")
    f.write("0.0 0.0 0.0\n")
    f.write("1.0 0.0 0.0\n")
    f.write("1.0 1.0 0.0\n")
    f.write("0.0 1.0 0.0\n")
    f.write("3 0 1 2\n")

ARGS = {
    "filepath": off_filepath,
    "points_only": True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, mpl_toolkits, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Mesh — a Mesh output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Mesh 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
options Ran
ran on a synthesized input
XiaoshuiHuang/fmr/evaluate.py
code served (permissive licence) · get_code("dd904972b5fe65ab")
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 Namespace. 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 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 argparse

ARGS = {
    'argv': [
        '-data', '7scene',
        '-o', './result/result.csv',
        '-p', './data/pert_010.csv',
        '-l', './result/log_010.log',
        '--pretrained', './result/fmr_model_7scene.pth',
        '--max-iter', '10',
        '--dim-k', '1024',
        '-j', '2',
        '--device', 'cpu',
        '-i', './data/7scene',
        '-c', './data/categories.txt',
        '--mode', 'test',
        '--uniformsampling', 'False'
    ]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Namespace — a Namespace output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Namespace 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
sinc1 Ran
ran on a synthesized input
XiaoshuiHuang/fmr/se_math/sinc.py
code served (permissive licence) · get_code("fa5ccc333d835363")
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)

# The function sinc1 takes a single parameter 't' which is a tensor
# It performs element-wise operations on the tensor, including:
# - torch.zeros_like(t)
# - torch.abs(t)
# - Comparison operations
# - Indexing with boolean masks
# - sin(t) and division operations
# So 't' should be a regular tensor (CPU, any numeric dtype)

t = torch.randn(2, 4)

ARGS = {
    "t": t
}
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
sinc1_dt Ran
ran on a synthesized input
XiaoshuiHuang/fmr/se_math/sinc.py
code served (permissive licence) · get_code("d6743b3739758b3c")
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)

# The function sinc1_dt takes a single parameter 't' which is used with:
# - torch.zeros_like(t): creates a tensor of zeros with same shape as t
# - torch.abs(t): absolute value
# - comparison operators (< and ==)
# - indexing with boolean masks (r[s], t[s], etc.)
# - squaring (t ** 2)
# - torch.cos() and torch.sin()
# - division operations
# So 't' should be a torch tensor (float type for math operations)

# Create a small tensor for 't' - batch size 2, with 4 elements each
t = torch.randn(2, 4, dtype=torch.float32)

ARGS = {
    "t": t
}
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
sinc1_dt_rt Ran
ran on a synthesized input
XiaoshuiHuang/fmr/se_math/sinc.py
code served (permissive licence) · get_code("2db19ceaf8b62ec1")
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 (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 tensor with values including some near zero and some larger
t = torch.tensor([0.001, 0.005, 0.5, 2.0], dtype=torch.float32)

ARGS = {
    "t": t
}
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 (4,) — shape (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
symfn_max Ran
ran on a synthesized input
XiaoshuiHuang/fmr/model.py
code served (permissive licence) · get_code("0531ab2ada61e54e")
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, 1) 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 tensor x and calls:
# - torch.nn.functional.max_pool1d(x, x.size(-1))
# This requires x to be a 3D tensor (batch, channels, length)
# for max_pool1d to work properly

# Create a small 3D tensor: (batch=2, channels=4, length=3)
x = torch.randn(2, 4, 3)

ARGS = {
    "x": x
}
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 (2, 4, 1) — shape (2, 4, 1) 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
batch_pinv_dx Not yet run
unverified (harvested, no recorded run)
XiaoshuiHuang/fmr/se_math/invmat.py
code served (permissive licence) · get_code("d764d841ff3bb5c9")
How it ran

no invocation recorded.

liebracket Not yet run
unverified (harvested, no recorded run)
XiaoshuiHuang/fmr/se_math/se3.py
code served (permissive licence) · get_code("a62cbc8f0b7428a4")
How it ran

no invocation recorded.

twist_prod Not yet run
unverified (harvested, no recorded run)
XiaoshuiHuang/fmr/se_math/se3.py
code served (permissive licence) · get_code("957002f51810e89d")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 14 are also present in other papers whose code we hold (14 with at least one paper dated earlier, 0 the same month, 0 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
batch_inverse 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
batch_inverse_dx 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
batch_pinv_dx 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
cross_prod 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
flatten 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
liebracket 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
mat 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
mat 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
objread 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
offread 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
sinc1 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
sinc1_dt 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
sinc1_dt_rt 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet
symfn_max 1 other paper
1 dated earlier
1903.05711 (2019-03)
PointNetLK: Robust &amp; Efficient Point Cloud Registration using PointNet

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

We present a fast feature-metric point cloud registration framework, which enforces the optimisation of registration by minimising a feature-metric projection error without correspondences. The advantage of the feature-metric projection error is robust to noise, outliers and density difference in contrast to the geometric projection error. Besides, minimising the feature-metric projection error does not need to search the correspondences so that the optimisation speed is fast. The principle behind the proposed method is that the feature difference is smallest if point clouds are aligned very well. We train the proposed method in a semi-supervised or unsupervised approach, which requires limited or no registration label data. Experiments demonstrate our method obtains higher accuracy and robustness than the state-of-the-art methods. Besides, experimental results show that the proposed method can handle significant noise and density difference, and solve both same-source and cross-source point cloud registration.

For agents

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

get_harvested_code_for_paper("2005.01014")
get_code_for_paper("2005.01014")
have("2005.01014")

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

curl https://syntology.ai/api/ran/2005.01014.json

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

[![Syntology run record](https://syntology.ai/api/ran/2005.01014.svg)](https://syntology.ai/paper/2005.01014)

Connect an agent — have() is free.