SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2308.01686 · ICCV · 2023

LiDAR-Camera Panoptic Segmentation via Geometry-Consistent and Semantic-Aware Alignment

Yuan Xie, Qian Yu, Zhiwei Zhang, Zhizhong Zhang, Ran Yi, Lizhuang Ma

arXiv · PDF · Open in the Atlas

Code that ran

9 of 11 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; 4 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 failures of our instrument; 2 no recorded run), measured 2026-11-22; 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/2308.01686.json · badge: /api/ran/2308.01686.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
zhangzw12319/lcps canonical (classifier label)
classifier basis: census
9 ran · 2 no recorded run
FunctionStatusWhere it lives
SemKITTI2train Ran
ran, honoured the contract we drafted
zhangzw12319/lcps/val.py
code served (permissive licence) · get_code("d1d458f2f7efe347")
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
import numpy as np

torch.manual_seed(0)

# The function SemKITTI2train takes a label parameter
# It can be either a single value or a list of values
# For each value, it subtracts 1

# Test case 1: single integer label
single_label = 5

# Test case 2: list of integer labels
list_labels = [10, 20, 15, 8]

# We'll test with a single label first
ARGS = {
    "label": single_label
}
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 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
SemKITTI2train_single Ran
ran, honoured the contract we drafted
zhangzw12319/lcps/val.py
code served (permissive licence) · get_code("b2573ee92b382019")
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 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

torch.manual_seed(0)

# The function SemKITTI2train_single takes a label and subtracts 1 from it.
# The label parameter should be a numeric value (int, float, or tensor).

label = torch.tensor(5)

ARGS = {
    "label": label
}
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 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
cart2polar Ran
ran, honoured the contract we drafted
zhangzw12319/lcps/dataloader/dataset.py
code served (permissive licence) · get_code("4cbfcba4a545864d")
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

# Create a small sample of 3D Cartesian coordinates (batch of 2, with x, y, z)
input_xyz = np.array([
    [1.0, 2.0, 3.0],
    [4.0, 5.0, 6.0]
], dtype=np.float32)

ARGS = {
    'input_xyz': input_xyz
}
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
conv3x3 Ran
ran; our contract draft was wrong, not the code
zhangzw12319/lcps/network/swiftnet.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from lixiaotong97/DSU, is pointer only (licence: NONE) · get_code("fac5364e2f53c6db")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from lixiaotong97/DSU, 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.

distributed_concat Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zhangzw12319/lcps/dataloader/eval_sampler.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from genjib/avsiam, is pointer only (licence: NONE) · get_code("fc419f8d3ed996b4")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from genjib/avsiam, 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.

grp_range_torch Ran
ran; our contract draft was wrong, not the code
zhangzw12319/lcps/network/ptBEV.py
code served (permissive licence) · get_code("bb6f31439534fc4b")
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 (6,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create a small tensor `a` for the group range function
# This should be a 1D tensor of positive integers representing group sizes
a = torch.tensor([2, 3, 1], dtype=torch.int64)

# Device should be CPU
dev = 'cpu'

ARGS = {
    'a': a,
    'dev': dev
}
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 (6,) — shape (6,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_pretrained_model Ran
ran; our contract draft was wrong, not the code
zhangzw12319/lcps/val.py
code served (permissive licence) · get_code("26d50efc5c59fc91")
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 SimpleModel. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Define a simple model with state_dict
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

# Create the main model
model = SimpleModel()

# Create a pretrained model state dict with some matching and non-matching keys
pretrained_model = {
    'linear1.weight': torch.randn(3, 4),
    'linear1.bias': torch.randn(3),
    'linear2.weight': torch.randn(2, 3),
    'extra_key_not_in_model': torch.randn(5, 5),  # This will be filtered out
}

ARGS = {
    'model': model,
    'pretrained_model': pretrained_model,
}
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 SimpleModel — a SimpleModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleModel 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
polar2cat Ran
ran, honoured the contract we drafted
zhangzw12319/lcps/dataloader/dataset.py
code served (permissive licence) · get_code("dd197502413d486a")
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, 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

# The function polar2cat takes input_xyz_polar which is indexed as:
# input_xyz_polar[0] - radius/rho component
# input_xyz_polar[1] - angle/theta component
# input_xyz_polar[2] - z component (height)
# These are used with np.cos, np.sin, and np.stack operations

# Create a simple input: array with shape (3,) or (3, batch_size)
# Using shape (3, 2) for batched polar coordinates
radius = np.array([1.0, 2.0])
angle = np.array([0.0, np.pi/4])
z = np.array([0.5, 1.5])

input_xyz_polar = np.stack((radius, angle, z), axis=0)

ARGS = {
    "input_xyz_polar": input_xyz_polar
}
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, 2) — shape (3, 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
return_tensor_index Ran
ran; our contract draft was wrong, not the code
zhangzw12319/lcps/network/pvp_generation.py
code served (permissive licence) · get_code("a9efda013e646584")
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 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

torch.manual_seed(0)

# Create a small 1D tensor to search for (M=4 elements)
value = torch.tensor([1.0, 2.0, 3.0, 4.0])

# Create a 2D tensor with N=2 vectors, each with M=4 elements
# First vector matches value, second is different
t = torch.tensor([
    [1.0, 2.0, 3.0, 4.0],
    [5.0, 6.0, 7.0, 8.0]
])

ARGS = {
    "value": value,
    "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 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
SwiftNetRes18 Not yet run
unverified (harvested, no recorded run)
zhangzw12319/lcps/network/swiftnet.py
code served (permissive licence) · get_code("acbf472f8220bcd4")
How it ran

no invocation recorded.

shift_voxel_grids Not yet run
unverified (harvested, no recorded run)
zhangzw12319/lcps/network/pvp_generation.py
code served (permissive licence) · get_code("9e9ab969e74a9fa8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 6 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 1 the same month, 3 later) and 5 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
conv3x3 686 other papers
498 dated earlier, 12 the same month, 169 later, 7 same year, order not determinable · 31 only as an assembly brick
1406.2661 (2014-06)
Generative Adversarial Networks
1409.0575 (2014-09)
ImageNet Large Scale Visual Recognition Challenge
1412.6572 (2014-12)
Explaining and Harnessing Adversarial Examples
and 683 more not named here
cart2polar 3 other papers
3 dated earlier
2003.14032 (2020-03)
PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation
2103.14962 (2021-03)
Panoptic-PolarNet: Proposal-free LiDAR Point Cloud Panoptic Segmentation
2109.05441 (2021-09)
Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR-based Perception
polar2cat 3 other papers
3 dated earlier
2003.14032 (2020-03)
PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation
2103.14962 (2021-03)
Panoptic-PolarNet: Proposal-free LiDAR Point Cloud Panoptic Segmentation
2109.05441 (2021-09)
Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR-based Perception
grp_range_torch 2 other papers
2 dated earlier
2003.14032 (2020-03)
PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation
2103.14962 (2021-03)
Panoptic-PolarNet: Proposal-free LiDAR Point Cloud Panoptic Segmentation
SemKITTI2train 1 other paper
1 later
2410.15674 (2024-10)
TALoS: Enhancing Semantic Scene Completion via Test-time Adaptation on the Line of Sight
distributed_concat 1 other paper
1 later
2403.19638 (2024-03)
Siamese Vision Transformers are Scalable Audio-visual Learners

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 panoptic segmentation is a challenging perception task that requires both semantic segmentation and instance segmentation. In this task, we notice that images could provide rich texture, color, and discriminative information, which can complement LiDAR data for evident performance improvement, but their fusion remains a challenging problem. To this end, we propose LCPS, the first LiDAR-Camera Panoptic Segmentation network. In our approach, we conduct LiDAR-Camera fusion in three stages: 1) an Asynchronous Compensation Pixel Alignment (ACPA) module that calibrates the coordinate misalignment caused by asynchronous problems between sensors; 2) a Semantic-Aware Region Alignment (SARA) module that extends the oneto-one point-pixel mapping to one-to-many semantic relations; 3) a Point-to-Voxel feature Propagation (PVP) module that integrates both geometric and semantic fusion information for the entire point cloud. Our fusion strategy improves about 6.9% PQ performance over the LiDAR-only baseline on NuScenes dataset. Extensive quantitative and qualitative experiments further demonstrate the effectiveness of our novel framework. The code will be released at https://github.com/zhangzw12319/lcps.git.

For agents

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

get_harvested_code_for_paper("2308.01686")
get_code_for_paper("2308.01686")
have("2308.01686")

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

curl https://syntology.ai/api/ran/2308.01686.json

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

[![Syntology run record](https://syntology.ai/api/ran/2308.01686.svg)](https://syntology.ai/paper/2308.01686)

Connect an agent — have() is free.