SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2109.09827 · ICCV · 2021

Viewpoint Invariant Dense Matching for Visual Geolocalization

Barbara Caputo, Gabriele Berton, Carlo Masone, Valerio Paolicelli

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 1 has 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/2109.09827.json · badge: /api/ran/2109.09827.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
gmberton/geo_warp marked official in Papers with Code
Papers with Code archive
3 ran · 0 no recorded run
FunctionStatusWhere it lives
Network Ran
ran on a synthesized input · constructed an object; no result was computed
gmberton/geo_warp/network.py
code served (permissive licence) · get_code("19f4b2268200acd4")
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 an instance of its own class, Network: the run constructed the object and did not call it, so no result was computed. 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 torch.nn as nn

torch.manual_seed(0)

# Stand-in FeaturesExtractor that outputs (B x 256 x 15 x 15)
class StandInFeaturesExtractor(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv = nn.Conv2d(3, 256, kernel_size=3, padding=1)
    
    def forward(self, tensor_images, features_type='local'):
        # Input: (2B x 3 x H x W), Output: (2B x 256 x 15 x 15)
        batch_size = tensor_images.shape[0]
        return torch.randn(batch_size, 256, 15, 15)

# Stand-in HomographyRegression that outputs (B x 16)
class StandInHomographyRegression(nn.Module):
    def __init__(self):
        super().__init__()
        self.fc = nn.Linear(225 * 15 * 15, 16)
    
    def forward(self, similarity_matrix):
        # Input: (B x 225 x 15 x 15), Output: (B x 16)
        batch_size = similarity_matrix.shape[0]
        flattened = similarity_matrix.view(batch_size, -1)
        return self.fc(flattened)

features_extractor = StandInFeaturesExtractor()
homography_regression = StandInHomographyRegression()

ARGS = {
    'features_extractor': features_extractor,
    'homography_regression': homography_regression
}
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 Network — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Network 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
compute_similarity Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
gmberton/geo_warp/network.py
code served (permissive licence) · get_code("f5a73718db5e20ab")
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 4, dtype family float, shape (2, 15, 3, 5) 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 torch.nn.functional as F

torch.manual_seed(0)

# Create tiny tensors with distinct dimensions
# Shape: (batch=2, channels=4, height=3, width=5)
features_a = torch.randn(2, 4, 3, 5)
features_b = torch.randn(2, 4, 3, 5)

ARGS = {
    "features_a": features_a,
    "features_b": features_b,
}
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 4, dtype family float, shape (2, 15, 3, 5) — shape (2, 15, 3, 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
feature_L2_norm Ran
ran, honoured the contract we drafted
gmberton/geo_warp/network.py
code served (permissive licence) · get_code("8cbff63e77c33ca2")
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)

# The function takes a feature tensor and normalizes it by L2 norm
# It expects a 2D tensor where dimension 0 is batch and dimension 1 is features
feature = torch.randn(2, 4)

ARGS = {
    "feature": feature
}
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 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 3 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.

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 paper we propose a novel method for image matching based on dense local features and tailored for visual geolocalization. Dense local features matching is robust against changes in illumination and occlusions, but not against viewpoint shifts which are a fundamental aspect of geolocalization. Our method, called GeoWarp, directly embeds invariance to viewpoint shifts in the process of extracting dense features. This is achieved via a trainable module which learns from the data an invariance that is meaningful for the task of recognizing places. We also devise a new self-supervised loss and two new weakly supervised losses to train this module using only unlabeled data and weak labels. GeoWarp is implemented efficiently as a re-ranking method that can be easily embedded into pre-existing visual geolocalization pipelines. Experimental validation on standard geolocalization benchmarks demonstrates that GeoWarp boosts the accuracy of state-of-the-art retrieval architectures. The code and trained models will be released upon acceptance of this paper.

For agents

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

get_harvested_code_for_paper("2109.09827")
get_code_for_paper("2109.09827")
have("2109.09827")

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

curl https://syntology.ai/api/ran/2109.09827.json

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

[![Syntology run record](https://syntology.ai/api/ran/2109.09827.svg)](https://syntology.ai/paper/2109.09827)

Connect an agent — have() is free.