SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.18511 · 2023

3DCoMPaT$^{++}$: An improved Large-scale 3D Vision Dataset for Compositional Recognition

arXiv · PDF · Open in the Atlas

Code that ran

11 of 16 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; 1 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 9 failures of our instrument; 5 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2310.18511.json · badge: /api/ran/2310.18511.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
cattalyya/3dcompat-challenge canonical (classifier label)
classifier basis: census
11 ran · 5 no recorded run
FunctionStatusWhere it lives
compute_overall_iou Ran
ran; our contract draft was wrong, not the code
cattalyya/3dcompat-challenge/models/3D/compat_utils.py
code served (permissive licence) · get_code("fc6a08422f9b2ca9")
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 small prediction and target numpy arrays
# Shape: (batch_size, height, width) with class labels
batch_size = 2
height = 4
width = 3
num_classes = 5

# Create prediction array with random class labels
np.random.seed(0)
pred_np = np.random.randint(0, num_classes, size=(batch_size, height, width))

# Create target array with random class labels
target_np = np.random.randint(0, num_classes, size=(batch_size, height, width))

ARGS = {
    "pred_np": pred_np,
    "target_np": target_np,
    "num_classes": num_classes,
}
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
load_json Ran
ran; our contract draft was wrong, not the code
cattalyya/3dcompat-challenge/download/download.py
code served (permissive licence) · get_code("7c9da204a931f831")
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 json
import tempfile
import os

# Create a temporary JSON file with valid content
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.json")

test_data = {"key": "value", "number": 42, "list": [1, 2, 3]}
with open(temp_file, 'w') as f:
    json.dump(test_data, f)

ARGS = {
    "file": temp_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
normalize_data Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cattalyya/3dcompat-challenge/models/3D/provider.py
code served (permissive licence) · get_code("4c642057e222b0f1")
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 3, dtype family float, shape (2, 4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create a small batch of point cloud data
# B=2 batches, N=4 points, C=3 coordinates
B, N, C = 2, 4, 3

# Generate sample point cloud data
batch_data = np.random.randn(B, N, C).astype(np.float32)

ARGS = {
    "batch_data": batch_data
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 4, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
open_hdf5 Ran
ran; our contract draft was wrong, not the code
cattalyya/3dcompat-challenge/models/2D_3D/feature_utils.py
code served (permissive licence) · get_code("f7588b28bd5e7304")
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 File. 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 h5py
import tempfile
import os

# Create a temporary HDF5 file for testing
temp_dir = tempfile.mkdtemp()
temp_hdf5_path = os.path.join(temp_dir, "test.h5")

# Create a valid HDF5 file first
with h5py.File(temp_hdf5_path, "w") as f:
    f.create_dataset("data", data=[1, 2, 3, 4])

ARGS = {
    "hdf5_file": temp_hdf5_path,
    "mode": "r"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
h5py — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type File — a File output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a File output the recorded ndim 1 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
parse_args Ran
ran; our contract draft was wrong, not the code
cattalyya/3dcompat-challenge/download/download.py
code served (permissive licence) · get_code("45e7c840ae8c6364")
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 Namespace. 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 argparse
import os
import tempfile

# Create a temporary directory for testing
temp_dir = tempfile.mkdtemp()

# Define ARGS dict with valid arguments for parse_args
ARGS = {
    'argv': [
        '--outdir', temp_dir,
        '--modality', '2D',
        '--split', 'train',
        '--semantic-level', 'fine',
        '--n-comp', 'compat_010'
    ]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Namespace — a Namespace output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Namespace 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
pc_normalize Ran
ran, honoured the contract we drafted
cattalyya/3dcompat-challenge/models/3D/compat_loader.py
code served (permissive licence) · get_code("7389309cb737203e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create a small point cloud: 4 points in 3D space
pc = np.array([
    [1.0, 2.0, 3.0],
    [4.0, 5.0, 6.0],
    [7.0, 8.0, 9.0],
    [10.0, 11.0, 12.0]
], dtype=np.float32)

ARGS = {
    "pc": pc
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (4, 3) — shape (4, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
read_npz Ran
ran; our contract draft was wrong, not the code
cattalyya/3dcompat-challenge/models/2D_3D/infer_submit.py
code served (permissive licence) · get_code("a6c19b923b5aa25b")
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 numpy as np
import tempfile
import os

# Create a temporary npz file with valid data structure
temp_dir = tempfile.mkdtemp()
temp_filepath = os.path.join(temp_dir, "test.npz")

# Create test data that matches what read_npz expects
# The function loads 'arr_0' and calls .item() on it, expecting a dict-like object
test_dict = {
    "key1": "value1",
    "key2": [1, 2, 3],
    "key3": np.array([4, 5, 6])
}

# Save as npz with the expected structure
np.savez(temp_filepath, test_dict)

ARGS = {
    "filepath": temp_filepath
}
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 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
shuffle_data Ran
ran; our contract draft was wrong, not the code
cattalyya/3dcompat-challenge/models/3D/provider.py
code served (permissive licence) · get_code("4794ca8cee394259")
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

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

# Create sample data and labels
# data: B,N,... numpy array where B is batch size
batch_size = 2
num_points = 4
feature_dim = 3

data = np.random.randn(batch_size, num_points, feature_dim)
labels = np.array([0, 1])

ARGS = {
    "data": data,
    "labels": labels,
}
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 — 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
shuffle_points Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cattalyya/3dcompat-challenge/models/3D/provider.py
code served (permissive licence) · get_code("00961a87304a9bf0")
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 3, dtype family float, shape (2, 4, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create a small batch of point cloud data
# Shape: BxNxC where B=batch size, N=num points, C=channels
batch_size = 2
num_points = 4
channels = 3

batch_data = np.random.randn(batch_size, num_points, channels).astype(np.float32)

ARGS = {
    "batch_data": batch_data
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 4, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
to_categorical Ran
ran; our contract draft was wrong, not the code
cattalyya/3dcompat-challenge/models/3D/compat_utils.py
code served (permissive licence) · get_code("66a55d4a270861e8")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (4, 5) as a later run recorded it. 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, 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 converts integer class labels to one-hot vectors
# Parameters:
# - y: a tensor of integer class indices
# - num_classes: an integer specifying the number of classes

# Create a small batch of class labels (integers from 0 to num_classes-1)
num_classes = 5
y = torch.tensor([0, 2, 4, 1], dtype=torch.long)

ARGS = {
    "y": y,
    "num_classes": num_classes,
}
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 (4, 5) — shape (4, 5) 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
valid_pixel Ran
ran, violated the contract we drafted
cattalyya/3dcompat-challenge/models/2D_3D/helper_2d3d.py
code served (permissive licence) · get_code("10d3bd8837459db9")
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 torch

torch.manual_seed(0)

# The function valid_pixel checks if coordinates (x, y) are valid within a square image of size imsize x imsize
# Parameters:
# - x: integer coordinate (horizontal)
# - y: integer coordinate (vertical)
# - imsize: integer representing the size of the square image

ARGS = {
    "x": 5,
    "y": 3,
    "imsize": 10
}
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 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
get_logits_from2d Not yet run
unverified (harvested, no recorded run)
cattalyya/3dcompat-challenge/models/2D_3D/helper_2d3d.py
code served (permissive licence) · get_code("0a94dc3d865b13d7")
How it ran

no invocation recorded.

inference Not yet run
unverified (harvested, no recorded run)
cattalyya/3dcompat-challenge/models/3D/infer_cls.py
code served (permissive licence) · get_code("365a5cbef894fb42")
How it ran

no invocation recorded.

inference Not yet run
unverified (harvested, no recorded run)
cattalyya/3dcompat-challenge/models/3D/infer_cls_submit.py
code served (permissive licence) · get_code("c3f4456346b94b58")
How it ran

no invocation recorded.

load_data Not yet run
unverified (harvested, no recorded run)
cattalyya/3dcompat-challenge/models/3D/compat_loader.py
code served (permissive licence) · get_code("711ea0b934da6b22")
How it ran

no invocation recorded.

project_to_2D Not yet run
unverified (harvested, no recorded run)
cattalyya/3dcompat-challenge/models/2D_3D/helper_2d3d.py
code served (permissive licence) · get_code("389349a926169aff")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 3 later) and 11 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
shuffle_data 17 other papers
15 dated earlier, 2 later
1803.05827 (2018-03)
Local Spectral Graph Convolution for Point Set Feature Learning
1803.11527 (2018-03)
SpiderCNN: Deep Learning on Point Sets with Parameterized Convolutional Filters
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
and 14 more not named here
shuffle_points 11 other papers
9 dated earlier, 2 later
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
1812.11647 (2018-12)
Path-Invariant Map Networks
2006.14865 (2020-06)
RPM-Net: Recurrent Prediction of Motion and Parts from Point Cloud
and 8 more not named here
normalize_data 6 other papers
4 dated earlier, 2 later
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
2103.16397 (2021-03)
3D AffordanceNet: A Benchmark for Visual Object Affordance Understanding
2109.09521 (2021-09)
RibSeg Dataset and Strong Point Cloud Baselines for Rib Segmentation from CT Scans
and 3 more not named here
to_categorical 4 other papers
4 dated earlier
1811.07246 (2018-11)
PointConv: Deep Convolutional Networks on 3D Point Clouds
2011.00931 (2020-11)
Point Transformer
2012.09688 (2020-12)
PCT: Point cloud transformer
and 1 more not named here
pc_normalize 1 other paper
1 dated earlier
2007.01294 (2020-07)
A Closer Look at Local Aggregation Operators in Point Cloud Analysis

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

Repositories linked to this paper

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

Abstract

In this work, we present 3DCoMPaT$^{++}$, a multimodal 2D/3D dataset with 160 million rendered views of more than 10 million stylized 3D shapes carefully annotated at the part-instance level, alongside matching RGB point clouds, 3D textured meshes, depth maps, and segmentation masks. 3DCoMPaT$^{++}$ covers 41 shape categories, 275 fine-grained part categories, and 293 fine-grained material classes that can be compositionally applied to parts of 3D objects. We render a subset of one million stylized shapes from four equally spaced views as well as four randomized views, leading to a total of 160 million renderings. Parts are segmented at the instance level, with coarse-grained and fine-grained semantic levels. We introduce a new task, called Grounded CoMPaT Recognition (GCR), to collectively recognize and ground compositions of materials on parts of 3D objects. Additionally, we report the outcomes of a data challenge organized at CVPR2023, showcasing the winning method's utilization of a modified PointNet$^{++}$ model trained on 6D inputs, and exploring alternative techniques for GCR enhancement. We hope our work will help ease future research on compositional 3D Vision.

For agents

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

get_harvested_code_for_paper("2310.18511")
get_code_for_paper("2310.18511")
have("2310.18511")

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

curl https://syntology.ai/api/ran/2310.18511.json

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

[![Syntology run record](https://syntology.ai/api/ran/2310.18511.svg)](https://syntology.ai/paper/2310.18511)

Connect an agent — have() is free.