SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.00619 · 2024

Harnessing Uncertainty-aware Bounding Boxes for Unsupervised 3D Object Detection

arXiv · PDF · Open in the Atlas

Code that ran

15 of 15 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; 6 honoured the contract we drafted, 3 violated it and 0 have no contract verdict; 6 failures of our instrument; 0 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/2408.00619.json · badge: /api/ran/2408.00619.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
Ruiyang-061X/UA3D canonical (classifier label)
classifier basis: census
15 ran · 0 no recorded run
FunctionStatusWhere it lives
box_to_string Ran
ran; our contract draft was wrong, not the code
Ruiyang-061X/UA3D/data_preprocessing/lyft/lyft2kitti.py
code served (permissive licence) · get_code("dd2865592c40bf95")
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 str. 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 numpy as np

class Box:
    """Minimal box class with required attributes for box_to_string"""
    def __init__(self):
        # rotation_matrix: used with np.dot to compute yaw
        self.rotation_matrix = np.array([
            [0.8, -0.6, 0.0],
            [0.6, 0.8, 0.0],
            [0.0, 0.0, 1.0]
        ])
        # wlh: width, length, height (accessed as wlh[2], wlh[0], wlh[1])
        self.wlh = np.array([5.2, 3.1, 1.8])
        # center: x, y, z coordinates
        self.center = np.array([10.5, 2.3, 0.9])

ARGS = {
    "name": "Car",
    "box": Box(),
    "bbox_2d": np.array([100.5, 150.3, 400.2, 450.8]),
    "truncation": 0.15,
    "occlusion": 1,
    "alpha": 0.52,
    "instance_token": "token_12345"
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
cart2hom Ran
ran, honoured the contract we drafted
Ruiyang-061X/UA3D/data_preprocessing/gather_historical_traversals.py
code served (permissive licence) · get_code("74ac280ac0823fc6")
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

# The function cart2hom converts 3D Cartesian coordinates to homogeneous coordinates
# by appending a column of ones to the input array.
# 
# Looking at the function body:
# - pts_3d.shape[0] is accessed (number of rows)
# - pts_3d is passed to np.hstack along with ones array
# - The function expects a 2D numpy array with shape (n, 3) for 3D points

# Create a small sample of 3D points (n=2, dim=3)
pts_3d = np.array([
    [1.0, 2.0, 3.0],
    [4.0, 5.0, 6.0]
], dtype=np.float32)

ARGS = {
    "pts_3d": pts_3d
}
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
cls_type_to_id Ran
ran, honoured the contract we drafted
Ruiyang-061X/UA3D/data_preprocessing/kitti_util.py
code served (permissive licence) · get_code("68d59bd1b88d6e20")
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 int, ndim 0, dtype family int, 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

ARGS = {
    'cls_type': 'Car'
}
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 int, ndim 0, dtype family int, 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
filter_by_ppscore Ran
ran; our contract draft was wrong, not the code
Ruiyang-061X/UA3D/generate_cluster_mask/combine_labels.py
code served (permissive licence) · get_code("83ab1e0e7536624e")
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 bool, 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 numpy as np

class MockObject:
    def __init__(self):
        self.ry = np.pi / 4  # rotation angle in radians
        self.l = 4.0  # length
        self.w = 2.0  # width
        self.h = 1.5  # height
        self.t = np.array([10.0, 5.0, 20.0])  # center position [x, y, z]

# Create sample point cloud in rectangular coordinates (x, y, z format)
ptc_rect = np.array([
    [9.0, 4.5, 19.5],
    [11.0, 5.0, 20.5],
    [10.5, 4.8, 20.2],
    [8.5, 5.2, 19.8],
], dtype=np.float32)

# Create pp_score array with values between 0 and 1
pp_score = np.array([0.3, 0.6, 0.7, 0.4], dtype=np.float32)

# Create object with required attributes
obj = MockObject()

ARGS = {
    'ptc_rect': ptc_rect,
    'pp_score': pp_score,
    'obj': obj,
    'percentile': 50,
    'threshold': 0.5,
}
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 bool, 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
find_closest_integer_in_ref_arr Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Ruiyang-061X/UA3D/data_preprocessing/nuscenes/nusc2kitti_boston.py
code served (permissive licence) · get_code("4974407b90a3a234")
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 int; 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 numpy as np

# Construct arguments for find_closest_integer_in_ref_arr
query_int = 42
ref_arr = np.array([10, 25, 50, 100, 200], dtype=np.int64)

ARGS = {
    "query_int": query_int,
    "ref_arr": ref_arr,
}
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 int — 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 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
in_hull Ran
ran, violated the contract we drafted
Ruiyang-061X/UA3D/generate_cluster_mask/pre_compute_pp_score.py
code served (permissive licence) · get_code("306616f38b29c8f3")
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 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 numpy as np
from scipy.spatial import Delaunay

# Create test points (N=2, K=3)
p = np.array([
    [0.5, 0.5, 0.5],
    [1.5, 1.5, 1.5]
], dtype=np.float64)

# Create hull corners (M=4, K=3) - a tetrahedron
hull = np.array([
    [0.0, 0.0, 0.0],
    [1.0, 0.0, 0.0],
    [0.0, 1.0, 0.0],
    [0.0, 0.0, 1.0]
], dtype=np.float64)

ARGS = {
    "p": p,
    "hull": hull
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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 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
in_hull Ran
ran, violated the contract we drafted
Ruiyang-061X/UA3D/generate_cluster_mask/gen_gt_mask.py
code served (permissive licence) · get_code("36c27fc61a6a9bb8")
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 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 numpy as np
from scipy.spatial import Delaunay

# Create test points (N=2, K=3)
p = np.array([
    [0.5, 0.5, 0.5],
    [1.5, 1.5, 1.5]
], dtype=np.float64)

# Create hull points forming a simple tetrahedron (M=4, K=3)
hull = np.array([
    [0.0, 0.0, 0.0],
    [1.0, 0.0, 0.0],
    [0.0, 1.0, 0.0],
    [0.0, 0.0, 1.0]
], dtype=np.float64)

ARGS = {
    "p": p,
    "hull": hull
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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 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
load_velo_scan Ran
ran, honoured the contract we drafted
Ruiyang-061X/UA3D/data_preprocessing/gather_historical_traversals.py
code served (permissive licence) · get_code("37c1e89b5864f71a")
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 (8, 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
import tempfile
import os

# Create a temporary binary file with synthetic velodyne scan data
temp_dir = tempfile.mkdtemp()
velo_filename = os.path.join(temp_dir, "scan.bin")

# Create synthetic velodyne scan data: N points with 4 values each (x, y, z, intensity)
# 8 points × 4 values = 32 float32 values
synthetic_scan = np.array([
    [1.0, 2.0, 3.0, 0.5],
    [4.0, 5.0, 6.0, 0.6],
    [7.0, 8.0, 9.0, 0.7],
    [10.0, 11.0, 12.0, 0.8],
    [13.0, 14.0, 15.0, 0.9],
    [16.0, 17.0, 18.0, 1.0],
    [19.0, 20.0, 21.0, 1.1],
    [22.0, 23.0, 24.0, 1.2],
], dtype=np.float32)

# Write to binary file
synthetic_scan.tofile(velo_filename)

ARGS = {
    "velo_filename": velo_filename
}
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 (8, 4) — shape (8, 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
postprocessing Ran
ran; our contract draft was wrong, not the code
Ruiyang-061X/UA3D/data_preprocessing/lyft/lyft2kitti.py
code served (permissive licence) · get_code("145ca2df8e53618f")
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 1, 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 numpy as np

# Create sample objects with bounding boxes and depths
objs = [
    {
        'bbox_2d': [10.0, 20.0, 50.0, 80.0],  # [x1, y1, x2, y2]
        'depth': 5.0
    },
    {
        'bbox_2d': [30.0, 40.0, 70.0, 90.0],
        'depth': 3.0
    },
    {
        'bbox_2d': [5.0, 5.0, 25.0, 35.0],
        'depth': 8.0
    }
]

height = 100
width = 120

ARGS = {
    'objs': objs,
    'height': height,
    'width': width
}
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 list, ndim 1, 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
predicts2objs Ran
ran; our contract draft was wrong, not the code
Ruiyang-061X/UA3D/generate_cluster_mask/combine_labels.py
code served (permissive licence) · get_code("748a64a37b298a5e")
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 1, 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 torch
from types import SimpleNamespace

torch.manual_seed(0)

# Construct preds dictionary with required keys
preds = {
    'location': torch.randn(2, 3),  # batch_size=2, 3D coordinates
    'dimensions': torch.randn(2, 3),  # batch_size=2, [length, height, width]
    'rotation_y': torch.randn(2),  # batch_size=2, rotation angles
    'score': torch.randn(2)  # batch_size=2, confidence scores
}

ARGS = {
    'preds': preds
}
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 list, ndim 1, 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
range_cutoff Ran
ran, violated the contract we drafted
Ruiyang-061X/UA3D/generate_cluster_mask/gen_gt_mask.py
code served (permissive licence) · get_code("7f710b9dd1564d49")
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

torch.manual_seed(0)

# The function accesses ptc[:, 0] and compares with 80
# So ptc needs to be a 2D tensor where the first column can be compared to 80
# Shape: (batch_size, num_features) where batch_size=2, num_features=4
ptc = torch.tensor([
    [50.0, 1.0, 2.0, 3.0],
    [90.0, 4.0, 5.0, 6.0]
])

ARGS = {
    "ptc": ptc
}
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
remove_center Ran
ran, honoured the contract we drafted
Ruiyang-061X/UA3D/generate_cluster_mask/pre_compute_pp_score.py
code served (permissive licence) · get_code("ebebb4d0ffe1438f")
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 (7, 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 numpy as np

# Create a point cloud with some points inside and outside the specified ranges
np.random.seed(0)
ptc = np.random.uniform(-2, 2, size=(10, 2)).astype(np.float32)

ARGS = {
    'ptc': ptc,
    'x_range': (-1.15, 1.75),
    'y_range': (-0.65, 0.65),
}
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 (7, 2) — shape (7, 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
rotx Ran
ran, honoured the contract we drafted
Ruiyang-061X/UA3D/data_preprocessing/kitti_util.py
code served (permissive licence) · get_code("8843d80ab666c4e9")
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 (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 numpy as np

def rotx(t):
    """ 3D Rotation about the x-axis. """
    c = np.cos(t)
    s = np.sin(t)
    return np.array([[1, 0, 0], [0, c, -s], [0, s, c]])

# Setup: t should be a scalar or array representing angle(s) in radians
t_value = np.pi / 4  # 45 degrees in radians

ARGS = {
    "t": t_value
}
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 (3, 3) — shape (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
roty Ran
ran, honoured the contract we drafted
Ruiyang-061X/UA3D/data_preprocessing/kitti_util.py
code served (permissive licence) · get_code("ae86074e10e1dfd7")
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 (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 numpy as np

# The function roty takes a single parameter 't' which represents an angle in radians
# The function computes a 3x3 rotation matrix about the y-axis
# The parameter should be a scalar (float or int) representing the rotation angle

t_value = np.pi / 4  # 45 degrees as a reasonable test value

ARGS = {
    "t": t_value
}
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 (3, 3) — shape (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
transform_points Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Ruiyang-061X/UA3D/data_preprocessing/gather_historical_traversals.py
code served (permissive licence) · get_code("8b80c6bbc9d282b1")
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, 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

# Construct arguments for transform_points function
# The function takes:
# - pts_3d_ref: 3D points array that gets converted to homogeneous coordinates
# - Tr: transformation matrix that is transposed and used in dot product

# Create sample 3D points (n points with 3 coordinates each)
# Using 2 points with 3D coordinates
pts_3d_ref = np.array([
    [1.0, 2.0, 3.0],
    [4.0, 5.0, 6.0]
], dtype=np.float32)

# Create transformation matrix
# After cart2hom, pts_3d becomes (n, 4)
# np.dot(pts_3d_ref, np.transpose(Tr)) should produce (n, 4)
# So Tr should be (4, 4) to make transpose (4, 4) and dot product (n, 4)
Tr = np.array([
    [1.0, 0.0, 0.0, 0.0],
    [0.0, 1.0, 0.0, 0.0],
    [0.0, 0.0, 1.0, 0.0],
    [0.1, 0.2, 0.3, 1.0]
], dtype=np.float32)

ARGS = {
    'pts_3d_ref': pts_3d_ref,
    'Tr': Tr
}
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, 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

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 15 are also present in other papers whose code we hold (15 with at least one paper dated earlier, 0 the same month, 0 later) and 0 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
roty 13 other papers
13 dated earlier
1612.07695 (2016-12)
MultiNet: Real-time Joint Semantic Reasoning for Autonomous Driving
1811.07782 (2018-11)
Modeling Local Geometric Structure of 3D Point Clouds using Geo-CNN
1912.12147 (2019-12)
Cooperative Perception for 3D Object Detection in Driving Scenarios using Infrastructure Sensors
and 10 more not named here
rotx 12 other papers
12 dated earlier
1612.07695 (2016-12)
MultiNet: Real-time Joint Semantic Reasoning for Autonomous Driving
1811.07782 (2018-11)
Modeling Local Geometric Structure of 3D Point Clouds using Geo-CNN
1912.12147 (2019-12)
Cooperative Perception for 3D Object Detection in Driving Scenarios using Infrastructure Sensors
and 9 more not named here
cls_type_to_id 6 other papers
6 dated earlier
2006.11275 (2020-06)
Center-based 3D Object Detection and Tracking
2007.11901 (2020-07)
Weakly Supervised 3D Object Detection from Lidar Point Cloud
2203.11405 (2022-03)
Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception
and 3 more not named here
cart2hom 4 other papers
4 dated earlier
2203.11405 (2022-03)
Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception
2203.15882 (2022-03)
Learning to Detect Mobile Objects from LiDAR Scans Without Labels
2303.15286 (2023-03)
Unsupervised Adaptation from Repeated Traversals for Autonomous Driving
and 1 more not named here
load_velo_scan 4 other papers
4 dated earlier
2203.11405 (2022-03)
Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception
2203.15882 (2022-03)
Learning to Detect Mobile Objects from LiDAR Scans Without Labels
2303.15286 (2023-03)
Unsupervised Adaptation from Repeated Traversals for Autonomous Driving
and 1 more not named here
transform_points 4 other papers
4 dated earlier
2203.11405 (2022-03)
Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception
2203.15882 (2022-03)
Learning to Detect Mobile Objects from LiDAR Scans Without Labels
2303.15286 (2023-03)
Unsupervised Adaptation from Repeated Traversals for Autonomous Driving
and 1 more not named here
box_to_string 3 other papers
3 dated earlier
2203.11405 (2022-03)
Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception
2203.15882 (2022-03)
Learning to Detect Mobile Objects from LiDAR Scans Without Labels
2303.15286 (2023-03)
Unsupervised Adaptation from Repeated Traversals for Autonomous Driving
in_hull 3 other papers
3 dated earlier
2203.15882 (2022-03)
Learning to Detect Mobile Objects from LiDAR Scans Without Labels
2303.15286 (2023-03)
Unsupervised Adaptation from Repeated Traversals for Autonomous Driving
2306.06023 (2023-06)
DetZero: Rethinking Offboard 3D Object Detection with Long-term Sequential Point Clouds
postprocessing 3 other papers
3 dated earlier
2203.11405 (2022-03)
Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception
2203.15882 (2022-03)
Learning to Detect Mobile Objects from LiDAR Scans Without Labels
2303.15286 (2023-03)
Unsupervised Adaptation from Repeated Traversals for Autonomous Driving
remove_center 2 other papers
2 dated earlier
2203.15882 (2022-03)
Learning to Detect Mobile Objects from LiDAR Scans Without Labels
2303.15286 (2023-03)
Unsupervised Adaptation from Repeated Traversals for Autonomous Driving
filter_by_ppscore 1 other paper
1 dated earlier
2203.15882 (2022-03)
Learning to Detect Mobile Objects from LiDAR Scans Without Labels
find_closest_integer_in_ref_arr 1 other paper
1 dated earlier
2203.11405 (2022-03)
Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception
in_hull 1 other paper
1 dated earlier
2203.15882 (2022-03)
Learning to Detect Mobile Objects from LiDAR Scans Without Labels
predicts2objs 1 other paper
1 dated earlier
2203.15882 (2022-03)
Learning to Detect Mobile Objects from LiDAR Scans Without Labels
range_cutoff 1 other paper
1 dated earlier
2203.15882 (2022-03)
Learning to Detect Mobile Objects from LiDAR Scans Without Labels

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

Unsupervised 3D object detection aims to identify objects of interest from unlabeled raw data, such as LiDAR points. Recent approaches usually adopt pseudo 3D bounding boxes (3D bboxes) from clustering algorithm to initialize the model training. However, pseudo bboxes inevitably contain noise, and such inaccuracies accumulate to the final model, compromising the performance. Therefore, in an attempt to mitigate the negative impact of inaccurate pseudo bboxes, we introduce a new uncertainty-aware framework for unsupervised 3D object detection, dubbed UA3D. In particular, our method consists of two phases: uncertainty estimation and uncertainty regularization. (1) In the uncertainty estimation phase, we incorporate an extra auxiliary detection branch alongside the original primary detector. The prediction disparity between the primary and auxiliary detectors could reflect fine-grained uncertainty at the box coordinate level. (2) Based on the assessed uncertainty, we adaptively adjust the weight of every 3D bbox coordinate via uncertainty regularization, refining the training process on pseudo bboxes. For pseudo bbox coordinate with high uncertainty, we assign a relatively low loss weight. Extensive experiments verify that the proposed method is robust against the noisy pseudo bboxes, yielding substantial improvements on nuScenes and Lyft compared to existing approaches, with increases of +6.9% AP$_{BEV}$ and +2.5% AP$_{3D}$ on nuScenes, and +4.1% AP$_{BEV}$ and +2.0% AP$_{3D}$ on Lyft.

For agents

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

get_harvested_code_for_paper("2408.00619")
get_code_for_paper("2408.00619")
have("2408.00619")

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

curl https://syntology.ai/api/ran/2408.00619.json

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

[![Syntology run record](https://syntology.ai/api/ran/2408.00619.svg)](https://syntology.ai/paper/2408.00619)

Connect an agent — have() is free.