SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2107.02583 · 2021

Point Cloud Registration using Representative Overlapping Points

arXiv · PDF · Open in the Atlas

Code that ran

4 of 9 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 4 have no contract verdict; 0 failures of our instrument; 5 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/2107.02583.json · badge: /api/ran/2107.02583.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
zhulf0804/ROPNet canonical (classifier label)
classifier basis: census
4 ran · 5 no recorded run
FunctionStatusWhere it lives
Ol_loss Ran
ran on a synthesized input
zhulf0804/ROPNet/src/loss/loss.py
code served (permissive licence) · get_code("887c1893371473e2")
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 torch.nn as nn

torch.manual_seed(0)

# x_ol and y_ol are used as inputs to CrossEntropyLoss
# CrossEntropyLoss expects logits of shape (batch_size, num_classes)
# so x_ol and y_ol should have shape (batch_size, num_classes)
batch_size = 2
num_classes = 4

x_ol = torch.randn(batch_size, num_classes)
y_ol = torch.randn(batch_size, num_classes)

# dists is used with torch.min(dists, dim=-1) and torch.min(dists, dim=1)
# This means dists needs at least 2 dimensions
# dim=-1 and dim=1 are used, so dists should have shape (batch_size, something)
# Let's make it (batch_size, 3) to be distinct from num_classes=4
dists = torch.randn(batch_size, 3)

ARGS = {
    "x_ol": x_ol,
    "y_ol": y_ol,
    "dists": dists
}
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 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
Refine_loss Ran
ran on a synthesized input
zhulf0804/ROPNet/src/loss/loss.py
code served (permissive licence) · get_code("b2558fe134681d23")
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

torch.manual_seed(0)

# Build pred_transformed_src: a list of tensors, each representing predictions at different iterations
# The function iterates over this list, so we need at least 2 elements
pred_transformed_src = [
    torch.randn(2, 4),  # batch_size=2, feature_dim=4
    torch.randn(2, 4),  # second iteration prediction
]

# Build gt_transformed_src: ground truth tensor
# Based on the function, this is compared against each element in pred_transformed_src
# The subtraction happens: pred_transformed_src[i] - gt_transformed_src
# So gt_transformed_src should have compatible shape (2, 4)
gt_transformed_src = torch.randn(2, 4)

# Build weights: optional weights tensor
# Looking at the weighted loss computation:
# weights * torch.mean(torch.abs(pred_transformed_src[i] - gt_transformed_src), dim=-1)
# The mean over dim=-1 reduces the last dimension, so result is shape (batch,)
# weights should be broadcastable to (batch,) or (batch, 1)
weights = torch.ones(2, 1)

# loss_type parameter is a string but not actually used in the function body
loss_type = 'mae'

ARGS = {
    'gt_transformed_src': gt_transformed_src,
    'pred_transformed_src': pred_transformed_src,
    'weights': weights,
    'loss_type': loss_type,
}
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 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
gather_points Ran
ran on a synthesized input
zhulf0804/ROPNet/src/models/model_utils.py
code served (permissive licence) · get_code("bf5a042b5e21fef5")
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, 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)

# Create small tensors for testing
B = 2  # batch size
N = 5  # number of points
C = 3  # number of channels
M = 3  # number of samples

# points: shape=(B, N, C)
points = torch.randn(B, N, C)

# inds: shape=(B, M) - indices to gather
inds = torch.randint(0, N, (B, M))

ARGS = {
    "points": points,
    "inds": inds,
}
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, 3, 3) — shape (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
summary_metrics Ran
ran on a synthesized input
zhulf0804/ROPNet/src/metrics/helper.py
code served (permissive licence) · get_code("a75b944d4b0c7723")
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, ndim 1, dtype family float; no output shape is stated. It ran in syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1) at 2026-09-27T00:55:41Z. If your copy returns the same type and ndim 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

# The function expects 6 parameters, each of which should be a list of numpy arrays
# that can be concatenated along axis=0. Each array should contain metric values.

# Create small numpy arrays for each metric type
# Using distinct sizes where appropriate for later probing

# r_mse: list of arrays with shape (batch, features)
r_mse = [
    np.random.rand(2, 3),
    np.random.rand(2, 3)
]

# r_mae: list of arrays with shape (batch, features)
r_mae = [
    np.random.rand(2, 3),
    np.random.rand(2, 3)
]

# t_mse: list of arrays with shape (batch, features)
t_mse = [
    np.random.rand(2, 4),
    np.random.rand(2, 4)
]

# t_mae: list of arrays with shape (batch, features)
t_mae = [
    np.random.rand(2, 4),
    np.random.rand(2, 4)
]

# r_isotropic: list of arrays with shape (batch, features)
r_isotropic = [
    np.random.rand(2, 3),
    np.random.rand(2, 3)
]

# t_isotropic: list of arrays with shape (batch, features)
t_isotropic = [
    np.random.rand(2, 4),
    np.random.rand(2, 4)
]

ARGS = {
    "r_mse": r_mse,
    "r_mae": r_mae,
    "t_mse": t_mse,
    "t_mae": t_mae,
    "r_isotropic": r_isotropic,
    "t_isotropic": t_isotropic
}
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, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
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 syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1); 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
Error_R Not yet run
unverified (harvested, no recorded run)
zhulf0804/ROPNet/src/metrics/metrics.py
code served (permissive licence) · get_code("f2a8fd97f5d5dd7d")
How it ran

no invocation recorded.

Error_t Not yet run
unverified (harvested, no recorded run)
zhulf0804/ROPNet/src/metrics/metrics.py
code served (permissive licence) · get_code("93b354421e2cda17")
How it ran

no invocation recorded.

Init_loss Not yet run
unverified (harvested, no recorded run)
zhulf0804/ROPNet/src/loss/loss.py
code served (permissive licence) · get_code("d1bb4076804a0c58")
How it ran

no invocation recorded.

anisotropic_R_error Not yet run
unverified (harvested, no recorded run)
zhulf0804/ROPNet/src/metrics/metrics.py
code served (permissive licence) · get_code("5a1b6afa12862dde")
How it ran

no invocation recorded.

pcd2npy Not yet run
unverified (harvested, no recorded run)
zhulf0804/ROPNet/src/utils/format.py
code served (permissive licence) · get_code("b82fcd19a698117e")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 9 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.

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

3D point cloud registration is a fundamental task in robotics and computer vision. Recently, many learning-based point cloud registration methods based on correspondences have emerged. However, these methods heavily rely on such correspondences and meet great challenges with partial overlap. In this paper, we propose ROPNet, a new deep learning model using Representative Overlapping Points with discriminative features for registration that transforms partial-to-partial registration into partial-to-complete registration. Specifically, we propose a context-guided module which uses an encoder to extract global features for predicting point overlap score. To better find representative overlapping points, we use the extracted global features for coarse alignment. Then, we introduce a Transformer to enrich point features and remove non-representative points based on point overlap score and feature matching. A similarity matrix is built in a partial-to-complete mode, and finally, weighted SVD is adopted to estimate a transformation matrix. Extensive experiments over ModelNet40 using noisy and partially overlapping point clouds show that the proposed method outperforms traditional and learning-based methods, achieving state-of-the-art performance. The code is available at https://github.com/zhulf0804/ROPNet.

For agents

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

get_harvested_code_for_paper("2107.02583")
get_code_for_paper("2107.02583")
have("2107.02583")

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

curl https://syntology.ai/api/ran/2107.02583.json

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

[![Syntology run record](https://syntology.ai/api/ran/2107.02583.svg)](https://syntology.ai/paper/2107.02583)

Connect an agent — have() is free.