SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1911.04460 · 2019

360SD-Net: 360° Stereo Depth Estimation with Learnable Cost Volume

arXiv · PDF · Open in the Atlas

Code that ran

9 of 10 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 6 have no contract verdict; 2 failures of our instrument; 1 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/1911.04460.json · badge: /api/ran/1911.04460.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
albert100121/360SD-Net community (classifier label)
classifier basis not recorded on the link
9 ran · 1 no recorded run
FunctionStatusWhere it lives
color_normalize Ran
ran on a synthesized input
albert100121/360SD-Net/dataloader/preprocess.py
code served (permissive licence) · get_code("547001586e1978ab")
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 Compose. 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 torch
import torchvision.transforms as transforms
from PIL import Image
import numpy as np

torch.manual_seed(0)

# The function color_normalize takes one parameter: normalize
# It expects normalize to be a dict with 'mean' and 'std' keys
# These are passed to transforms.Normalize(**normalize)

# Create a custom normalize dict
custom_normalize = {
    'mean': [0.5, 0.5, 0.5],
    'std': [0.2, 0.2, 0.2]
}

ARGS = {
    'normalize': custom_normalize
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Compose — a Compose output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Compose 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
color_preproccess Ran
ran on a synthesized input
albert100121/360SD-Net/dataloader/preprocess.py
code served (permissive licence) · get_code("8981d3f0947d85a0")
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 Compose. 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 torch
import torchvision.transforms as transforms
from PIL import Image
import numpy as np

torch.manual_seed(0)

# Create a small test image as PIL Image
# The function expects PIL Images as input to the transform pipeline
test_image_array = np.random.randint(0, 256, (32, 32, 3), dtype=np.uint8)
test_image = Image.fromarray(test_image_array)

# Define the normalize parameter
__imagenet_stats = {'mean': [0.485, 0.456, 0.406], 'std': [0.229, 0.224, 0.225]}

ARGS = {
    'normalize': __imagenet_stats
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Compose — a Compose output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Compose 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
convbn Ran
ran; our contract draft was wrong, not the code
albert100121/360SD-Net/models/sub_ASPP.py
pointer only (licence: MIT); the stored copy get_code serves, from 3dcvdeveloper/bgnet, is pointer only (licence: NONE) · get_code("15ab3c76823aef7e")
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 3dcvdeveloper/bgnet, 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.

convbn_3d Ran
ran; our contract draft was wrong, not the code
albert100121/360SD-Net/models/sub_ASPP.py
pointer only (licence: MIT); the stored copy get_code serves, from mickeykang16/temporaleventstereo, is pointer only (licence: NONE) · get_code("46674adf87c5459d")
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 mickeykang16/temporaleventstereo, 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.

dataloader Ran
ran on a synthesized input
albert100121/360SD-Net/dataloader/filename_loader.py
code served (permissive licence) · get_code("2f98e3ffc934fb62")
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, ndim 2, 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 os
import tempfile
import shutil

# Create temporary directories with the required structure
temp_dir = tempfile.mkdtemp()
val_dir = tempfile.mkdtemp()

# Create subdirectories
os.makedirs(os.path.join(temp_dir, 'image_up'), exist_ok=True)
os.makedirs(os.path.join(temp_dir, 'image_down'), exist_ok=True)
os.makedirs(os.path.join(temp_dir, 'disp_up'), exist_ok=True)

os.makedirs(os.path.join(val_dir, 'image_up'), exist_ok=True)
os.makedirs(os.path.join(val_dir, 'image_down'), exist_ok=True)
os.makedirs(os.path.join(val_dir, 'disp_up'), exist_ok=True)

# Create dummy files for training data
for i in range(2):
    # Create .png files
    open(os.path.join(temp_dir, 'image_up', f'img_{i}.png'), 'w').close()
    open(os.path.join(temp_dir, 'image_down', f'img_{i}.png'), 'w').close()
    # Create .npy files
    open(os.path.join(temp_dir, 'disp_up', f'img_{i}.npy'), 'w').close()

# Create dummy files for validation data
for i in range(2):
    # Create .png files
    open(os.path.join(val_dir, 'image_up', f'val_{i}.png'), 'w').close()
    open(os.path.join(val_dir, 'image_down', f'val_{i}.png'), 'w').close()
    # Create .npy files
    open(os.path.join(val_dir, 'disp_up', f'val_{i}.npy'), 'w').close()

ARGS = {
    'filepath': temp_dir + '/',
    'valpath': val_dir + '/'
}
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 tuple, ndim 2, dtype family float — a tuple 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
dataloader Ran
ran on a synthesized input
albert100121/360SD-Net/dataloader/testing_loader.py
code served (permissive licence) · get_code("c244e29f33dfb60c")
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, ndim 2, 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 os
import tempfile
import shutil

# Create a temporary directory structure with image files
temp_dir = tempfile.mkdtemp()
up_fold_path = os.path.join(temp_dir, 'image_up')
down_fold_path = os.path.join(temp_dir, 'image_down')

os.makedirs(up_fold_path, exist_ok=True)
os.makedirs(down_fold_path, exist_ok=True)

# Create dummy image files
for i in range(2):
    open(os.path.join(up_fold_path, f'image_{i}.png'), 'w').close()
    open(os.path.join(down_fold_path, f'image_{i}.png'), 'w').close()

# Also add a non-image file to test filtering
open(os.path.join(up_fold_path, 'readme.txt'), 'w').close()

ARGS = {
    'filepath': temp_dir + '/'
}
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 tuple, ndim 2, dtype family float — a tuple 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
default_loader Ran
ran, honoured the contract we drafted
albert100121/360SD-Net/dataloader/RGB_Loader.py
pointer only (licence: MIT); the stored copy get_code serves, from xlgsyzp/cvpt, is pointer only (licence: NONE) · get_code("ac269a0e4b8d946e")
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 xlgsyzp/cvpt, 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.

disparity_loader Ran
ran on a synthesized input
albert100121/360SD-Net/dataloader/RGB_Loader.py
code served (permissive licence) · get_code("65d90cc99b34e88e")
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
import tempfile
import os

# Create a temporary directory and a dummy .npy file
temp_dir = tempfile.mkdtemp()
dummy_array = np.random.rand(4, 3).astype(np.float32)
npy_path = os.path.join(temp_dir, "disparity.npy")
np.save(npy_path, dummy_array)

ARGS = {
    "path": npy_path
}
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
get_transform Ran
ran on a synthesized input
albert100121/360SD-Net/dataloader/preprocess.py
code served (permissive licence) · get_code("229c631eafaeffd7")
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 Compose. 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 torch
import torchvision.transforms as transforms

torch.manual_seed(0)

__imagenet_stats = {'mean': [0.485, 0.456, 0.406], 'std': [0.229, 0.224, 0.225]}

ARGS = {
    'name': 'imagenet',
    'normalize': None,
    'augment': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Compose — a Compose output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Compose 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
default_loader Not yet run
unverified (harvested, no recorded run)
albert100121/360SD-Net/dataloader/grayscale_Loader.py
code served (permissive licence) · get_code("069eeaeeefea7573")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 2 the same month, 3 later) and 7 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
default_loader 27 other papers
6 dated earlier, 21 later
1710.03463 (2017-10)
Learning to Generalize: Meta-Learning for Domain Generalization
1803.08669 (2018-03)
Pyramid Stereo Matching Network
1810.11408 (2018-10)
Anytime Stereo Image Depth Estimation on Mobile Devices
and 24 more not named here
convbn 13 other papers
3 dated earlier, 1 the same month, 9 later
1803.08669 (2018-03)
Pyramid Stereo Matching Network
1806.01260 (2018-06)
Digging Into Self-Supervised Monocular Depth Estimation
1905.00538 (2019-05)
DPSNet: End-to-end Deep Plane Sweep Stereo
and 10 more not named here
convbn_3d 9 other papers
3 dated earlier, 1 the same month, 5 later
1803.08669 (2018-03)
Pyramid Stereo Matching Network
1806.01260 (2018-06)
Digging Into Self-Supervised Monocular Depth Estimation
1905.00538 (2019-05)
DPSNet: End-to-end Deep Plane Sweep Stereo
and 6 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

Recently, end-to-end trainable deep neural networks have significantly improved stereo depth estimation for perspective images. However, 360° images captured under equirectangular projection cannot benefit from directly adopting existing methods due to distortion introduced (i.e., lines in 3D are not projected onto lines in 2D). To tackle this issue, we present a novel architecture specifically designed for spherical disparity using the setting of top-bottom 360° camera pairs. Moreover, we propose to mitigate the distortion issue by (1) an additional input branch capturing the position and relation of each pixel in the spherical coordinate, and (2) a cost volume built upon a learnable shifting filter. Due to the lack of 360° stereo data, we collect two 360° stereo datasets from Matterport3D and Stanford3D for training and evaluation. Extensive experiments and ablation study are provided to validate our method against existing algorithms. Finally, we show promising results on real-world environments capturing images with two consumer-level cameras.

For agents

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

get_harvested_code_for_paper("1911.04460")
get_code_for_paper("1911.04460")
have("1911.04460")

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

curl https://syntology.ai/api/ran/1911.04460.json

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

[![Syntology run record](https://syntology.ai/api/ran/1911.04460.svg)](https://syntology.ai/paper/1911.04460)

Connect an agent — have() is free.