SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1904.02113 · 2019

Point Cloud Oversegmentation with Graph-Structured Deep Metric Learning

arXiv · PDF · Open in the Atlas

Code that ran

13 of 13 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 12 have no contract verdict; 1 failure 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/1904.02113.json · badge: /api/ran/1904.02113.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
loicland/superpoint_graph canonical (classifier label)
classifier basis: census
12 ran · 0 no recorded run
jsgaobiao/superpoint_graph community (classifier label)
classifier basis: plan_row
1 ran · 0 no recorded run
FunctionStatusWhere it lives
compute_boundary_precision Ran
ran on a synthesized input
loicland/superpoint_graph/learning/metrics.py
code served (permissive licence) · get_code("53525019df188c6b")
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 numpy as np

torch.manual_seed(0)

# Analyze the function:
# - is_transition: used with == comparison and * operation, needs to be a tensor/array of booleans
# - pred_transitions: used with == comparison, * operation, and .sum() method, needs to be a tensor/array of booleans
# - Both should have the same shape for element-wise operations
# - The result is a scalar (sum divided by sum)

# Create small boolean tensors
is_transition = torch.tensor([True, False, True, False], dtype=torch.bool)
pred_transitions = torch.tensor([True, False, False, False], dtype=torch.bool)

ARGS = {
    "is_transition": is_transition,
    "pred_transitions": pred_transitions,
}
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 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
compute_boundary_recall Ran
ran on a synthesized input
loicland/superpoint_graph/learning/metrics.py
code served (permissive licence) · get_code("3cff4b0c303f38be")
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)

# The function compute_boundary_recall expects:
# - is_transition: a tensor of boolean/binary values indicating true transitions
# - pred_transitions: a tensor of boolean/binary values indicating predicted transitions
# Both should have the same shape and support element-wise comparison and sum operations

is_transition = torch.tensor([1, 0, 1, 0], dtype=torch.float32)
pred_transitions = torch.tensor([1, 0, 1, 1], dtype=torch.float32)

ARGS = {
    "is_transition": is_transition,
    "pred_transitions": pred_transitions,
}
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
compute_dist Ran
ran on a synthesized input
loicland/superpoint_graph/supervized_partition/losses.py
code served (permissive licence) · get_code("2782cfa4e987b0a1")
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 (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 small embeddings tensor (num_nodes=5, embedding_dim=4)
embeddings = torch.randn(5, 4, dtype=torch.float32)

# Create edge indices (source and target node indices)
# Let's create 3 edges
edg_source = torch.tensor([0, 1, 2], dtype=torch.long)
edg_target = torch.tensor([1, 2, 3], dtype=torch.long)

# Normalize embeddings for intrinsic and scalar distance types to avoid numerical issues
embeddings = embeddings / (torch.norm(embeddings, dim=1, keepdim=True) + 1e-8)

# Valid dist_type values: 'euclidian', 'intrinsic', or 'scalar'
dist_type = 'euclidian'

ARGS = {
    'embeddings': embeddings,
    'edg_source': edg_source,
    'edg_target': edg_target,
    'dist_type': dist_type,
}
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 1, dtype family float, 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
compute_loss Ran
ran on a synthesized input
loicland/superpoint_graph/supervized_partition/losses.py
code served (permissive licence) · get_code("851ed5d8a57bf8c7")
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. 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

torch.manual_seed(0)

# Create a minimal args class with required attributes
class Args:
    def __init__(self, loss, dist_type):
        self.loss = loss
        self.dist_type = dist_type

# Test with 'tv' and 'zhang' loss
args = Args(loss='tv_zhang', dist_type='euclidian')

# Create small tensors
# diff: differences between edges (shape: batch_size,)
diff = torch.tensor([0.5, 0.3, 0.8, 0.2], dtype=torch.float32)

# is_transition: binary indicator for transition (0=intra, 1=inter)
is_transition = torch.tensor([0, 0, 1, 1], dtype=torch.long)

# weights_loss: weights for each edge
weights_loss = torch.tensor([0.1, 0.2, 0.3, 0.4], dtype=torch.float32)

ARGS = {
    'args': args,
    'diff': diff,
    'is_transition': is_transition,
    'weights_loss': weights_loss
}
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 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 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
compute_predicted_transitions Ran
ran on a synthesized input
loicland/superpoint_graph/learning/metrics.py
code served (permissive licence) · get_code("f6312733479ecec8")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 1, dtype family float, shape (4,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Analyzing the function:
# - in_component: indexed with edg_source and edg_target, supports != comparison
# - edg_source: used as index into in_component
# - edg_target: used as index into in_component
# The function compares elements at different indices, so these should be array-like

# Create in_component as a 1D tensor/array that can be indexed
in_component = torch.tensor([0, 1, 0, 1, 2], dtype=torch.long)

# Create edge source and target indices (small, distinct values)
edg_source = torch.tensor([0, 1, 2, 3], dtype=torch.long)
edg_target = torch.tensor([1, 2, 3, 4], dtype=torch.long)

ARGS = {
    "in_component": in_component,
    "edg_source": edg_source,
    "edg_target": edg_target,
}
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 1, dtype family float, shape (4,) — shape (4,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
create_fnet Ran
ran; our contract draft was wrong, not the code
jsgaobiao/superpoint_graph/learning/graphnet.py
pointer only (licence: MIT); the stored copy get_code serves, from MMCheng/SSPC-Net, is pointer only (licence: NONE) · get_code("e7175798629e58c2")
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 MMCheng/SSPC-Net, 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.

get_info Ran
ran on a synthesized input
loicland/superpoint_graph/learning/custom_dataset.py
code served (permissive licence) · get_code("787baadda400bd94")
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 dict. 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 numpy as np

class Args:
    def __init__(self, pc_attribs, edge_attribs):
        self.pc_attribs = pc_attribs
        self.edge_attribs = edge_attribs

ARGS = {
    'args': Args(
        pc_attribs='x,y,z,nx,ny,nz,curvature,intensity,label',
        edge_attribs='delta_avg,delta_std,xyz,distance'
    )
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
get_info Ran
ran on a synthesized input
loicland/superpoint_graph/learning/s3dis_dataset.py
code served (permissive licence) · get_code("ba1f19c5139c4825")
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 dict. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

class Args:
    def __init__(self):
        self.edge_attribs = 'delta_avg,delta_std,xyz,feat1'
        self.loss_weights = 'none'
        self.S3DIS_PATH = '/fake/path'
        self.cvfold = 1
        self.cuda = False
        self.pc_attribs = 'x,y,z,intensity'

ARGS = {
    'args': Args()
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
h5py, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
get_info Ran
ran on a synthesized input
loicland/superpoint_graph/learning/sema3d_dataset.py
code served (permissive licence) · get_code("b36f9b676c19b53b")
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 dict. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

class Args:
    def __init__(self):
        self.edge_attribs = 'delta_avg,delta_std,xyz,scalar1'
        self.pc_attribs = 'feat1,feat2,feat3'
        self.loss_weights = 'none'
        self.SEMA3D_PATH = '/tmp'
        self.cuda = False

ARGS = {'args': Args()}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
h5py, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
get_s3dis_info Ran
ran on a synthesized input
loicland/superpoint_graph/supervized_partition/graph_processing.py
code served (permissive licence) · get_code("834ef8633c9081cb")
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 dict. 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

# The function get_s3dis_info only takes 'args' as a parameter
# Looking at the function body, it doesn't actually use 'args' at all
# It just returns a hardcoded dictionary with classes and inv_class_map
# So 'args' can be any object - we'll use a simple dict or namespace

class SimpleArgs:
    pass

args = SimpleArgs()

ARGS = {
    'args': args
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
get_sema3d_info Ran
ran on a synthesized input
loicland/superpoint_graph/supervized_partition/graph_processing.py
code served (permissive licence) · get_code("60d50c0625cc0aa7")
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 dict. 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)

class Args:
    def __init__(self):
        pass

ARGS = {
    'args': Args()
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
read_structure Ran
ran on a synthesized input
loicland/superpoint_graph/supervized_partition/graph_processing.py
code served (permissive licence) · get_code("cbecc43af2c952d3")
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. 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
import h5py
import tempfile
import os

# Create a temporary HDF5 file with the required datasets
temp_dir = tempfile.mkdtemp()
file_path = os.path.join(temp_dir, "test_structure.h5")

with h5py.File(file_path, 'w') as f:
    # Create datasets with small but distinct dimensions
    f.create_dataset('xyz', data=np.random.randn(10, 3).astype('float32'))
    f.create_dataset('rgb', data=np.random.randn(10, 3).astype('float32'))
    f.create_dataset('elevation', data=np.random.randn(10, 1).astype('float32'))
    f.create_dataset('xyn', data=np.random.randn(10, 2).astype('float32'))
    
    # Edge indices
    f.create_dataset('source', data=np.array([0, 1, 2, 3], dtype='int'))
    f.create_dataset('target', data=np.array([1, 2, 3, 4], dtype='int'))
    
    # Transition labels
    f.create_dataset('is_transition', data=np.array([0, 1, 0, 1], dtype='int'))
    
    # Objects
    f.create_dataset('objects', data=np.array(['obj1', 'obj2'], dtype=object))
    
    # Labels
    f.create_dataset('labels', data=np.array([0, 1, 0, 1], dtype='int'))
    
    # Geometry features
    f.create_dataset('geof', data=np.random.randn(4, 4).astype('float32'))
    f.create_dataset('target_local_geometry', data=np.random.randint(0, 100, (4, 4), dtype='uint32'))

ARGS = {
    'file_name': file_path,
    'read_geof': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
h5py, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
zhang Ran
ran on a synthesized input
loicland/superpoint_graph/supervized_partition/losses.py
code served (permissive licence) · get_code("b08b6b16d4684b7b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Construct arguments for zhang function
x = torch.randn(2, 4)  # Small tensor: batch size 2, dimension 4
lam = 0.5  # Scalar lambda parameter
dist_type = 'euclidian'  # Distance type string

ARGS = {
    'x': x,
    'lam': lam,
    'dist_type': dist_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 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

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 12 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
create_fnet 1 other paper
1 later
2104.07861 (2021-04)
SSPC-Net: Semi-supervised Semantic 3D Point Cloud Segmentation Network

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We propose a new supervized learning framework for oversegmenting 3D point clouds into superpoints. We cast this problem as learning deep embeddings of the local geometry and radiometry of 3D points, such that the border of objects presents high contrasts. The embeddings are computed using a lightweight neural network operating on the points' local neighborhood. Finally, we formulate point cloud oversegmentation as a graph partition problem with respect to the learned embeddings. This new approach allows us to set a new state-of-the-art in point cloud oversegmentation by a significant margin, on a dense indoor dataset (S3DIS) and a sparse outdoor one (vKITTI). Our best solution requires over five times fewer superpoints to reach similar performance than previously published methods on S3DIS. Furthermore, we show that our framework can be used to improve superpoint-based semantic segmentation algorithms, setting a new state-of-the-art for this task as well.

For agents

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

get_harvested_code_for_paper("1904.02113")
get_code_for_paper("1904.02113")
have("1904.02113")

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

curl https://syntology.ai/api/ran/1904.02113.json

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

[![Syntology run record](https://syntology.ai/api/ran/1904.02113.svg)](https://syntology.ai/paper/1904.02113)

Connect an agent — have() is free.