SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2003.03164 · 2020

D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features

arXiv · PDF · Open in the Atlas

Code that ran

6 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; 1 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 3 failures of our instrument; 2 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/2003.03164.json · badge: /api/ran/2003.03164.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
XuyangBai/D3Feat canonical (classifier label)
classifier basis: census
6 ran · 2 no recorded run
FunctionStatusWhere it lives
all_diffs Ran
ran, honoured the contract we drafted
XuyangBai/D3Feat/utils/loss.py
code served (permissive licence) · get_code("8427c7e156becb53")
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 EagerTensor, ndim 3, dtype family float, shape (2, 3, 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 tensorflow as tf

# Create two 2D tensors for testing
# a: batch size 2, feature dimension 4
a = tf.constant([[1.0, 2.0, 3.0, 4.0],
                 [5.0, 6.0, 7.0, 8.0]], dtype=tf.float32)

# b: batch size 3, feature dimension 4
b = tf.constant([[0.5, 1.0, 1.5, 2.0],
                 [2.5, 3.0, 3.5, 4.0],
                 [4.5, 5.0, 5.5, 6.0]], dtype=tf.float32)

ARGS = {
    'a': a,
    'b': b
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 3, dtype family float, shape (2, 3, 4) — shape (2, 3, 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
bias_variable Ran
ran; our contract draft was wrong, not the code
XuyangBai/D3Feat/models/network_blocks.py
code served (permissive licence) · get_code("6a88f1ae24fa26b2")
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 ResourceVariable, 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 tensorflow as tf

def bias_variable(shape):
    initial = tf.constant(0.0, shape=shape)
    return tf.Variable(initial, name='bias')

# The function takes a single parameter 'shape' which should be a tuple or list
# defining the shape of the bias variable. We'll create a small shape tuple.

ARGS = {
    'shape': (4,)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ResourceVariable, 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
cdist Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
XuyangBai/D3Feat/utils/loss.py
code served (permissive licence) · get_code("3dbcae6ca6415ede")
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 EagerTensor, 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 tensorflow as tf

# Create small 2D tensors with distinct dimensions
# a: (B1=2, F=4), b: (B2=3, F=4)
a = tf.constant([
    [1.0, 2.0, 3.0, 4.0],
    [5.0, 6.0, 7.0, 8.0]
], dtype=tf.float32)

b = tf.constant([
    [1.5, 2.5, 3.5, 4.5],
    [5.5, 6.5, 7.5, 8.5],
    [0.5, 1.5, 2.5, 3.5]
], dtype=tf.float32)

metric = 'euclidean'

ARGS = {
    'a': a,
    'b': b,
    'metric': metric
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, 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
get_at_indices Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
XuyangBai/D3Feat/utils/loss.py
code served (permissive licence) · get_code("f6fa452e61831f1d")
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 EagerTensor, 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 tensorflow as tf
import numpy as np

# Create a 2D tensor with shape (2, 4) - batch size 2, 4 features
tensor = tf.constant([[1.0, 2.0, 3.0, 4.0],
                      [5.0, 6.0, 7.0, 8.0]])

# Create indices tensor with shape (2,) - one index per batch element
# Valid indices should be in range [0, 3] since tensor has 4 columns
indices = tf.constant([0, 2], dtype=tf.int32)

ARGS = {
    'tensor': tensor,
    'indices': indices
}
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 EagerTensor, 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
rotate Ran
ran on a synthesized input
XuyangBai/D3Feat/datasets/ThreeDMatch.py
code served (permissive licence) · get_code("8d97dff7e23e80ea")
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

# The function takes:
# - points: a numpy array of shape (N, 3) representing 3D points
# - num_axis: an integer specifying rotation mode (1 or 3)

# Create a small set of 3D points
points = np.random.randn(2, 3).astype(np.float32)

ARGS = {
    "points": points,
    "num_axis": 1
}
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
rotate Ran
ran on a synthesized input
XuyangBai/D3Feat/datasets/ThreeDMatch_back.py
code served (permissive licence) · get_code("5af2ecb7d64ab693")
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: shape (4, 3) for 4 points in 3D space
points = np.random.randn(4, 3).astype(np.float32)

ARGS = {
    "points": points,
    "num_axis": 1
}
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
ind_max_pool Not yet run
unverified (harvested, no recorded run)
XuyangBai/D3Feat/models/network_blocks.py
code served (permissive licence) · get_code("4383ecd8cc0719fa")
How it ran

no invocation recorded.

weight_variable Not yet run
unverified (harvested, no recorded run)
XuyangBai/D3Feat/models/network_blocks.py
code served (permissive licence) · get_code("5a6188c7d118c3fc")
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, 6 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 3 the same month, 6 later) and 2 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
bias_variable 4 other papers
1 dated earlier, 1 the same month, 2 later
1904.08889 (2019-04)
KPConv: Flexible and Deformable Convolution for Point Clouds
2003.13035 (2020-03)
Multi-Path Region Mining For Weakly Supervised 3D Semantic Segmentation on Point Clouds
2007.06888 (2020-07)
JSENet: Joint Semantic Segmentation and Edge Detection Network for 3D Point Clouds
and 1 more not named here
ind_max_pool 4 other papers
1 dated earlier, 1 the same month, 2 later
1904.08889 (2019-04)
KPConv: Flexible and Deformable Convolution for Point Clouds
2003.13035 (2020-03)
Multi-Path Region Mining For Weakly Supervised 3D Semantic Segmentation on Point Clouds
2007.06888 (2020-07)
JSENet: Joint Semantic Segmentation and Edge Detection Network for 3D Point Clouds
and 1 more not named here
weight_variable 4 other papers
1 dated earlier, 1 the same month, 2 later
1904.08889 (2019-04)
KPConv: Flexible and Deformable Convolution for Point Clouds
2003.13035 (2020-03)
Multi-Path Region Mining For Weakly Supervised 3D Semantic Segmentation on Point Clouds
2007.06888 (2020-07)
JSENet: Joint Semantic Segmentation and Edge Detection Network for 3D Point Clouds
and 1 more not named here
all_diffs 3 other papers
2 dated earlier, 1 later
1703.07737 (2017-03)
In Defense of the Triplet Loss for Person Re-Identification
1811.06879 (2018-11)
The Perfect Match: 3D Point Cloud Matching with Smoothed Densities
2203.14493 (2022-03)
ARCS: Accurate Rotation and Correspondence Search
cdist 3 other papers
2 dated earlier, 1 later
1703.07737 (2017-03)
In Defense of the Triplet Loss for Person Re-Identification
1811.06879 (2018-11)
The Perfect Match: 3D Point Cloud Matching with Smoothed Densities
2203.14493 (2022-03)
ARCS: Accurate Rotation and Correspondence Search
get_at_indices 3 other papers
2 dated earlier, 1 later
1703.07737 (2017-03)
In Defense of the Triplet Loss for Person Re-Identification
1811.06879 (2018-11)
The Perfect Match: 3D Point Cloud Matching with Smoothed Densities
2203.14493 (2022-03)
ARCS: Accurate Rotation and Correspondence Search

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

A successful point cloud registration often lies on robust establishment of sparse matches through discriminative 3D local features. Despite the fast evolution of learning-based 3D feature descriptors, little attention has been drawn to the learning of 3D feature detectors, even less for a joint learning of the two tasks. In this paper, we leverage a 3D fully convolutional network for 3D point clouds, and propose a novel and practical learning mechanism that densely predicts both a detection score and a description feature for each 3D point. In particular, we propose a keypoint selection strategy that overcomes the inherent density variations of 3D point clouds, and further propose a self-supervised detector loss guided by the on-the-fly feature matching results during training. Finally, our method achieves state-of-the-art results in both indoor and outdoor scenarios, evaluated on 3DMatch and KITTI datasets, and shows its strong generalization ability on the ETH dataset. Towards practical use, we show that by adopting a reliable feature detector, sampling a smaller number of features is sufficient to achieve accurate and fast point cloud alignment.[code release](https://github.com/XuyangBai/D3Feat)

For agents

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

get_harvested_code_for_paper("2003.03164")
get_code_for_paper("2003.03164")
have("2003.03164")

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

curl https://syntology.ai/api/ran/2003.03164.json

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

[![Syntology run record](https://syntology.ai/api/ran/2003.03164.svg)](https://syntology.ai/paper/2003.03164)

Connect an agent — have() is free.