SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2304.03420 · 2023

Toward Unsupervised 3D Point Cloud Anomaly Detection using Variational Autoencoder

arXiv · PDF · Open in the Atlas

Code that ran

9 of 14 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 8 have no contract verdict; 1 failure 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/2304.03420.json · badge: /api/ran/2304.03420.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
llien30/point_cloud_anomaly_detection canonical (classifier label)
classifier basis not recorded on the link
9 ran · 5 no recorded run
FunctionStatusWhere it lives
change2positive Ran
ran on a synthesized input
llien30/point_cloud_anomaly_detection/libs/dataset.py
code served (permissive licence) · get_code("3051d9785f3880b3")
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 2, dtype family float, shape (4, 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

# The function takes a pointcloud as input
# It expects a numpy array with shape (N, 3) where N is number of points
# and the 3 columns are x, y, z coordinates

# Create a small pointcloud with some negative values
pointcloud = np.array([
    [-5.0, -3.0, 2.0],
    [1.0, 2.0, -1.0],
    [3.0, -2.0, 4.0],
    [0.5, 1.5, 0.5]
], dtype=np.float32)

ARGS = {
    "pointcloud": pointcloud
}
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 2, dtype family float, shape (4, 3) — shape (4, 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
fartherst_point_sampling Ran
ran on a synthesized input
llien30/point_cloud_anomaly_detection/libs/sampling.py
code served (permissive licence) · get_code("dc34e8a17f1354d5")
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 1, dtype family int, shape (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

np.random.seed(0)

# Create a small point cloud: 6 points with 3 coordinates each
points = np.array([
    [0.0, 0.0, 0.0],
    [1.0, 0.0, 0.0],
    [0.0, 1.0, 0.0],
    [1.0, 1.0, 0.0],
    [0.5, 0.5, 0.5],
    [2.0, 2.0, 2.0]
], dtype=np.float32)

def l2_norm(x, y):
    """Calculate l2 norm (distance) of `x` and `y`.
    Args:
        x (numpy.ndarray or cupy): (batch_size, num_point, coord_dim)
        y (numpy.ndarray): (batch_size, num_point, coord_dim)
    Returns (numpy.ndarray): (batch_size, num_point)
    """
    return ((x - y) ** 2).sum(axis=1)

ARGS = {
    'points': points,
    'num_sample_point': 3,
    'initial_idx': None,
    'metrics': l2_norm,
    'indices_dtype': np.int32,
    'distances_dtype': np.float32
}
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 1, dtype family int, shape (3,) — shape (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
knn Ran
ran on a synthesized input
llien30/point_cloud_anomaly_detection/libs/foldingnet.py
code served (permissive licence) · get_code("09f5929f369fcf3e")
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 int, shape (16,) 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)

# Construct a small 3D tensor: (batch_size=2, num_features=3, num_points=4)
x = torch.randn(2, 3, 4)

# k parameter for KNN
k = 2

ARGS = {
    "x": x,
    "k": k,
}
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 int, shape (16,) — shape (16,) 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
l2_norm Ran
ran on a synthesized input
llien30/point_cloud_anomaly_detection/libs/sampling.py
code served (permissive licence) · get_code("1c009741e97fbc84")
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 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 numpy as np

# Set random seed for reproducibility
np.random.seed(0)

# Create small numpy arrays with distinct dimensions
# batch_size=2, num_point=3, coord_dim=4
x = np.random.randn(2, 3, 4).astype(np.float32)
y = np.random.randn(2, 3, 4).astype(np.float32)

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
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, 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
loadOBJ Ran
ran on a synthesized input
llien30/point_cloud_anomaly_detection/libs/load_obj.py
code served (permissive licence) · get_code("919de3f905b5d468")
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 list, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. 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 tempfile
import os

# Create a temporary OBJ file with valid vertex data
temp_dir = tempfile.gettempdir()
temp_filepath = os.path.join(temp_dir, "test_model.obj")

with open(temp_filepath, 'w') as f:
    f.write("# Test OBJ file\n")
    f.write("v 1.0 2.0 3.0\n")
    f.write("v 4.0 5.0 6.0\n")
    f.write("v 7.0 8.0 9.0\n")
    f.write("# Some comment\n")
    f.write("v 10.0 11.0 12.0\n")

ARGS = {
    "filepath": temp_filepath
}
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 list, ndim 2, dtype family float — a list 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 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
local_maxpool Ran
ran on a synthesized input
llien30/point_cloud_anomaly_detection/libs/foldingnet.py
code served (permissive licence) · get_code("b734b82b1586447a")
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

torch.manual_seed(0)

# Create a small input tensor
# The function expects x to have shape (batch_size, channels, num_points)
# Let's use batch_size=2, channels=4, num_points=3
batch_size = 2
num_points = 3
channels = 4

x = torch.randn(batch_size, channels, num_points)

# Create idx - this is used to index into the flattened tensor
# x gets reshaped to (batch_size * num_points, -1) which is (6, 4)
# So idx should be valid indices into this, e.g., a tensor of indices
# Looking at the function, idx is used as: x[idx, :]
# This suggests idx should be a 1D tensor of indices
idx = torch.tensor([0, 1, 2, 3, 4, 5])

ARGS = {
    "x": x,
    "idx": idx
}
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, 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
mse_loss Ran
ran on a synthesized input
llien30/point_cloud_anomaly_detection/libs/loss.py
code served (permissive licence) · get_code("3ea6f5d1d3d84bb4")
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)

# Create small tensors for input and output
# Using distinct sizes: batch_size=2, feature_dim=4
input_tensor = torch.randn(2, 4)
output_tensor = torch.randn(2, 4)

ARGS = {
    'input': input_tensor,
    'output': output_tensor,
    'size_average': True
}
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
translate_pointcloud Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
llien30/point_cloud_anomaly_detection/libs/dataset.py
pointer only (licence: MIT); the stored copy get_code serves, from MosyMosy/Purge-Gate, is pointer only (licence: NONE) · get_code("791051c3e72b67d2")
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 MosyMosy/Purge-Gate, 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.

uniform_size Ran
ran on a synthesized input
llien30/point_cloud_anomaly_detection/libs/dataset.py
code served (permissive licence) · get_code("7c583bf08d7167d5")
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 (4, 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 pointcloud tensor with shape (4, 3)
# 4 points, 3 coordinates (x, y, z)
pointcloud = torch.tensor([
    [1.0, 2.0, 3.0],
    [4.0, 5.0, 6.0],
    [2.0, 3.0, 4.0],
    [5.0, 6.0, 7.0]
], dtype=torch.float32)

ARGS = {
    'pointcloud': pointcloud
}
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 2, dtype family float, shape (4, 3) — shape (4, 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
evaluate_foldingnet Not yet run
unverified (harvested, no recorded run)
llien30/point_cloud_anomaly_detection/libs/helper.py
code served (permissive licence) · get_code("7c3c0dd515969590")
How it ran

no invocation recorded.

local_cov Not yet run
unverified (harvested, no recorded run)
llien30/point_cloud_anomaly_detection/libs/foldingnet.py
code served (permissive licence) · get_code("0ec577ec5e304ba0")
How it ran

no invocation recorded.

resume Not yet run
unverified (harvested, no recorded run)
llien30/point_cloud_anomaly_detection/libs/checkpoint.py
code served (permissive licence) · get_code("9c220e3432a84bfa")
How it ran

no invocation recorded.

train_foldingnet Not yet run
unverified (harvested, no recorded run)
llien30/point_cloud_anomaly_detection/libs/helper.py
code served (permissive licence) · get_code("910faf785357a14a")
How it ran

no invocation recorded.

train_variational_foldingnet Not yet run
unverified (harvested, no recorded run)
llien30/point_cloud_anomaly_detection/libs/helper.py
code served (permissive licence) · get_code("1859040164445d63")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 13 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
translate_pointcloud 13 other papers
7 dated earlier, 6 later
1806.01411 (2018-06)
FlowNet3D: Learning Scene Flow in 3D Point Clouds
1906.12320 (2019-06)
PointFlow: 3D Point Cloud Generation with Continuous Normalizing Flows
1912.00195 (2019-12)
SGAS: Sequential Greedy Architecture Search
and 10 more not named here

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 paper, we present an end-to-end unsupervised anomaly detection framework for 3D point clouds. To the best of our knowledge, this is the first work to tackle the anomaly detection task on a general object represented by a 3D point cloud. We propose a deep variational autoencoder-based unsupervised anomaly detection network adapted to the 3D point cloud and an anomaly score specifically for 3D point clouds. To verify the effectiveness of the model, we conducted extensive experiments on the ShapeNet dataset. Through quantitative and qualitative evaluation, we demonstrate that the proposed method outperforms the baseline method. Our code is available at https://github.com/llien30/point_cloud_anomaly_detection.

For agents

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

get_harvested_code_for_paper("2304.03420")
get_code_for_paper("2304.03420")
have("2304.03420")

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

curl https://syntology.ai/api/ran/2304.03420.json

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

[![Syntology run record](https://syntology.ai/api/ran/2304.03420.svg)](https://syntology.ai/paper/2304.03420)

Connect an agent — have() is free.