SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1812.03320 · 2018

GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloud

arXiv · PDF · Open in the Atlas

Code that ran

4 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 0 failures of our instrument; 4 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/1812.03320.json · badge: /api/ran/1812.03320.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
ericyi/GSPN canonical (classifier label)
classifier basis: census
4 ran · 4 no recorded run
FunctionStatusWhere it lives
nn_distance_cpu Ran
ran on a synthesized input
ericyi/GSPN/tf_ops/nn_distance/tf_nndistance_cpu.py
code served (permissive licence) · get_code("6b16fa3eaa500e90")
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 tensorflow as tf
import numpy as np

# Set seeds for reproducibility
np.random.seed(0)
tf.random.set_seed(0)

# Create small point clouds
# pc1: (B=2, N=3, C=4) - batch of 2, 3 points each, 4 coordinates
pc1_data = np.random.randn(2, 3, 4).astype(np.float32)
pc1 = tf.constant(pc1_data)

# pc2: (B=2, M=5, C=4) - batch of 2, 5 points each, 4 coordinates
pc2_data = np.random.randn(2, 5, 4).astype(np.float32)
pc2 = tf.constant(pc2_data)

ARGS = {
    "pc1": pc1,
    "pc2": pc2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — 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
read_color_ply Ran
ran on a synthesized input
ericyi/GSPN/utils/io_util.py
code served (permissive licence) · get_code("3675ed8a249631ec")
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 (5, 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 numpy as np
import tempfile
import os
from plyfile import PlyData, PlyElement

# Create a temporary PLY file with color data
temp_dir = tempfile.mkdtemp()
temp_ply_file = os.path.join(temp_dir, "test_color.ply")

# Create sample point cloud data with colors
num_points = 5
x = np.array([0.1, 0.2, 0.3, 0.4, 0.5], dtype=np.float32)
y = np.array([1.1, 1.2, 1.3, 1.4, 1.5], dtype=np.float32)
z = np.array([2.1, 2.2, 2.3, 2.4, 2.5], dtype=np.float32)
red = np.array([255, 128, 64, 32, 16], dtype=np.uint8)
green = np.array([200, 150, 100, 50, 25], dtype=np.uint8)
blue = np.array([100, 75, 50, 25, 10], dtype=np.uint8)

# Create structured array for PLY
vertex_data = np.zeros(num_points, dtype=[('x', 'f4'), ('y', 'f4'), ('z', 'f4'), 
                                           ('red', 'u1'), ('green', 'u1'), ('blue', 'u1')])
vertex_data['x'] = x
vertex_data['y'] = y
vertex_data['z'] = z
vertex_data['red'] = red
vertex_data['green'] = green
vertex_data['blue'] = blue

# Create PLY element and file
vertex_element = PlyElement.describe(vertex_data, 'vertex')
ply_data = PlyData([vertex_element])
ply_data.write(temp_ply_file)

ARGS = {
    'filename': temp_ply_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, plyfile — 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 (5, 6) — shape (5, 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
read_label_ply Ran
ran on a synthesized input
ericyi/GSPN/utils/io_util.py
code served (permissive licence) · get_code("2cece9fb38995274")
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
from plyfile import PlyData, PlyElement
import tempfile
import os

# Create a temporary PLY file with the required structure
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.ply")

# Create sample data
n_points = 5
x = np.array([0.1, 0.2, 0.3, 0.4, 0.5], dtype=np.float32)
y = np.array([1.1, 1.2, 1.3, 1.4, 1.5], dtype=np.float32)
z = np.array([2.1, 2.2, 2.3, 2.4, 2.5], dtype=np.float32)
label = np.array([0, 1, 2, 1, 0], dtype=np.int32)

# Create vertex data structured array
vertex_data = np.zeros(n_points, dtype=[('x', 'f4'), ('y', 'f4'), ('z', 'f4'), ('label', 'i4')])
vertex_data['x'] = x
vertex_data['y'] = y
vertex_data['z'] = z
vertex_data['label'] = label

# Create PLY element and file
vertex = PlyElement.describe(vertex_data, 'vertex')
PlyData([vertex]).write(temp_file)

ARGS = {
    "filename": temp_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, plyfile — 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
read_ply Ran
ran on a synthesized input
ericyi/GSPN/utils/io_util.py
code served (permissive licence) · get_code("62046c0483d24f55")
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
from plyfile import PlyData, PlyElement
import tempfile
import os

# Create a temporary PLY file with valid point cloud data
temp_dir = tempfile.mkdtemp()
temp_ply_path = os.path.join(temp_dir, "test_cloud.ply")

# Create simple point cloud data
x = np.array([0.0, 1.0, 2.0], dtype=np.float32)
y = np.array([0.5, 1.5, 2.5], dtype=np.float32)
z = np.array([0.1, 1.1, 2.1], dtype=np.float32)

# Create vertex array for PLY file
vertex = np.array([(x[i], y[i], z[i]) for i in range(len(x))],
                  dtype=[('x', 'f4'), ('y', 'f4'), ('z', 'f4')])

# Write PLY file
el = PlyElement.describe(vertex, 'vertex')
PlyData([el]).write(temp_ply_path)

ARGS = {
    "filename": temp_ply_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, plyfile — 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
conv1d Not yet run
unverified (harvested, no recorded run)
ericyi/GSPN/utils/tf_util.py
code served (permissive licence) · get_code("798f68da5db33b27")
How it ran

no invocation recorded.

conv2d Not yet run
unverified (harvested, no recorded run)
ericyi/GSPN/utils/tf_util.py
code served (permissive licence) · get_code("7dc68c12bdd381e4")
How it ran

no invocation recorded.

conv2d_transpose Not yet run
unverified (harvested, no recorded run)
ericyi/GSPN/utils/tf_util.py
code served (permissive licence) · get_code("901a0f27987b65ef")
How it ran

no invocation recorded.

sample_and_group_all Not yet run
unverified (harvested, no recorded run)
ericyi/GSPN/utils/pointnet_util.py
code served (permissive licence) · get_code("142a56400c483118")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 3 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
sample_and_group_all 7 other papers
2 dated earlier, 1 the same month, 4 later
1803.05827 (2018-03)
Local Spectral Graph Convolution for Point Set Feature Learning
1811.07782 (2018-11)
Modeling Local Geometric Structure of 3D Point Clouds using Geo-CNN
1812.11647 (2018-12)
Path-Invariant Map Networks
and 4 more not named here
conv1d 5 other papers
1 dated earlier, 2 the same month, 2 later
1811.07782 (2018-11)
Modeling Local Geometric Structure of 3D Point Clouds using Geo-CNN
1812.00020 (2018-12)
TextureNet: Consistent Local Parametrizations for Learning from High-Resolution Signals on Meshes
1812.11647 (2018-12)
Path-Invariant Map Networks
and 2 more not named here
conv2d 5 other papers
1 dated earlier, 2 the same month, 2 later
1811.07782 (2018-11)
Modeling Local Geometric Structure of 3D Point Clouds using Geo-CNN
1812.00020 (2018-12)
TextureNet: Consistent Local Parametrizations for Learning from High-Resolution Signals on Meshes
1812.11647 (2018-12)
Path-Invariant Map Networks
and 2 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

We introduce a novel 3D object proposal approach named Generative Shape Proposal Network (GSPN) for instance segmentation in point cloud data. Instead of treating object proposal as a direct bounding box regression problem, we take an analysis-by-synthesis strategy and generate proposals by reconstructing shapes from noisy observations in a scene. We incorporate GSPN into a novel 3D instance segmentation framework named Region-based PointNet (R-PointNet) which allows flexible proposal refinement and instance segmentation generation. We achieve state-of-the-art performance on several 3D instance segmentation tasks. The success of GSPN largely comes from its emphasis on geometric understandings during object proposal, which greatly reducing proposals with low objectness.

For agents

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

get_harvested_code_for_paper("1812.03320")
get_code_for_paper("1812.03320")
have("1812.03320")

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

curl https://syntology.ai/api/ran/1812.03320.json

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

[![Syntology run record](https://syntology.ai/api/ran/1812.03320.svg)](https://syntology.ai/paper/1812.03320)

Connect an agent — have() is free.