SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2004.13324 · 2020

Learning Feature Descriptors using Camera Pose Supervision

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 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; 2 failures of our instrument; 1 no recorded run), measured 2026-11-21; 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/2004.13324.json · badge: /api/ran/2004.13324.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
qianqianwang68/caps community (classifier label)
classifier basis not recorded on the link
6 ran · 1 no recorded run
FunctionStatusWhere it lives
average_precision Ran
ran on a synthesized input
qianqianwang68/caps/utils.py
code served (permissive licence) · get_code("26f04718282dc148")
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 float, 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 numpy as np
import torch

torch.manual_seed(0)

# The function expects:
# - labels: num_examples x num_pts (binary labels)
# - logits: num_examples x num_pts (prediction scores)

num_examples = 2
num_pts = 4

# Create sample labels (binary, with at least one example having positive labels)
labels = np.array([
    [1, 0, 1, 0],
    [0, 0, 0, 0]
])

# Create sample logits (continuous scores)
logits = np.array([
    [0.9, 0.2, 0.8, 0.1],
    [0.5, 0.4, 0.6, 0.3]
])

ARGS = {
    "labels": labels,
    "logits": logits
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, 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
class_for_name Ran
ran; our contract draft was wrong, not the code
qianqianwang68/caps/CAPS/network.py
code served (permissive licence) · get_code("22e670c8093b7c80")
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 type. 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 importlib

torch_imported = False
try:
    import torch
    torch.manual_seed(0)
    torch_imported = True
except ImportError:
    pass

# The function class_for_name takes two string parameters:
# - module_name: the name of a module to import
# - class_name: the name of a class to retrieve from that module

# We'll use a real module and class that exist in Python's standard library
# For example, we can use the 'collections' module and get 'OrderedDict'

ARGS = {
    "module_name": "collections",
    "class_name": "OrderedDict"
}
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 type — a type output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a type 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
homogenize Ran
ran on a synthesized input
qianqianwang68/caps/utils.py
code served (permissive licence) · get_code("961908c17d9b6d4c")
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, 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 sample keypoint array with shape (n, 2)
# where n=5 keypoints, each with 2D coordinates (x, y)
kp = np.array([
    [10.5, 20.3],
    [15.2, 25.8],
    [30.1, 40.5],
    [5.9, 10.2],
    [50.0, 60.0]
], dtype=np.float32)

ARGS = {
    'kp': kp
}
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 (5, 3) — shape (5, 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
perspective_transform Ran
ran on a synthesized input
qianqianwang68/caps/dataloader/data_utils.py
code served (permissive licence) · get_code("c91cb9ba15e376f9")
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 cv2

# Create a small test image (4x3 for distinct dimensions)
# Using 3 channels (BGR) as typical for cv2 operations
img = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)

# Define the arguments dictionary
ARGS = {
    'img': img,
    'param': 0.001
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, 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
rotateImage Ran
ran on a synthesized input
qianqianwang68/caps/dataloader/data_utils.py
code served (permissive licence) · get_code("100f94f2543eec22")
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 cv2

# Create a small test image (height=100, width=150, 3 channels for color)
image = np.random.randint(0, 256, (100, 150, 3), dtype=np.uint8)

# Define rotation angle in degrees
angle = 30

ARGS = {
    "image": image,
    "angle": angle,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, 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
skew Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
qianqianwang68/caps/dataloader/data_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from KU-CVLAB/CoPoNeRF, is pointer only (licence: NONE) · get_code("6bd09766d7ad212b")
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 KU-CVLAB/CoPoNeRF, 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.

evaluate_pose Not yet run
unverified (harvested, no recorded run)
qianqianwang68/caps/utils.py
code served (permissive licence) · get_code("3ffdd49cf30a8dce")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 2 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
class_for_name 5 other papers
1 dated earlier, 4 later
1505.04597 (2015-05)
U-Net: Convolutional Networks for Biomedical Image Segmentation
2306.06359 (2023-06)
NeRFool: Uncovering the Vulnerability of Generalizable Neural Radiance Fields against Adversarial Perturbations
2308.11793 (2023-08)
Enhancing NeRF akin to Enhancing LLMs: Generalizable NeRF Transformer with Mixture-of-View-Experts
and 2 more not named here
skew 1 other paper
1 later
2312.07246 (2023-12)
Unifying Correspondence, Pose and NeRF for Pose-Free Novel View Synthesis from Stereo Pairs

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Recent research on learned visual descriptors has shown promising improvements in correspondence estimation, a key component of many 3D vision tasks. However, existing descriptor learning frameworks typically require ground-truth correspondences between feature points for training, which are challenging to acquire at scale. In this paper we propose a novel weakly-supervised framework that can learn feature descriptors solely from relative camera poses between images. To do so, we devise both a new loss function that exploits the epipolar constraint given by camera poses, and a new model architecture that makes the whole pipeline differentiable and efficient. Because we no longer need pixel-level ground-truth correspondences, our framework opens up the possibility of training on much larger and more diverse datasets for better and unbiased descriptors. We call the resulting descriptors CAmera Pose Supervised, or CAPS, descriptors. Though trained with weak supervision, CAPS descriptors outperform even prior fully-supervised descriptors and achieve state-of-the-art performance on a variety of geometric tasks. Project Page: https://qianqianwang68.github.io/CAPS/

For agents

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

get_harvested_code_for_paper("2004.13324")
get_code_for_paper("2004.13324")
have("2004.13324")

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

curl https://syntology.ai/api/ran/2004.13324.json

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

[![Syntology run record](https://syntology.ai/api/ran/2004.13324.svg)](https://syntology.ai/paper/2004.13324)

Connect an agent — have() is free.