SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.03453 · CVPR · 2023

OSRT: Omnidirectional Image Super-Resolution with Distortion-aware Transformer

Gen Li, Ying Shan, Xintao Wang, Chao Dong, Fanghua Yu, Mingdeng Cao

arXiv · PDF · Open in the Atlas

Code that ran

11 of 12 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; 2 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 6 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/2302.03453.json · badge: /api/ran/2302.03453.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
fanghua-yu/osrt canonical (classifier label)
classifier basis not recorded on the link
11 ran · 1 no recorded run
FunctionStatusWhere it lives
default_conv Ran
ran on a synthesized input
fanghua-yu/osrt/odisr/archs/launet_arch.py
code served (permissive licence) · get_code("0e1c2934ef064fde")
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 Conv2d. 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 torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "in_channels": 3,
    "out_channels": 64,
    "kernel_size": 3,
    "bias": True,
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
drop_path Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
fanghua-yu/osrt/odisr/archs/osrt_arch.py
pointer only (licence: MIT); the stored copy get_code serves, from jingyunliang/vrt, is pointer only (licence: NOASSERTION) · get_code("9ea4d5b8b4d5ef2f")
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 jingyunliang/vrt, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

erp_downsample_fisheye Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
fanghua-yu/osrt/odisr/utils/erp_downsample.py
code served (permissive licence) · get_code("5e1d924a740460bd")
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 3, dtype family float, shape (3, 8, 16) 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 numpy as np

torch.manual_seed(0)

# Create a small ERP HR tensor: [c, h, w]
# Using small dimensions: 3 channels, 16 height, 32 width
erp_hr = torch.rand(3, 16, 32)

# Define arguments for erp_downsample_fisheye function
ARGS = {
    'erp_hr': erp_hr,
    'scale': 2,
    'downsample_type': 'bicubic',
    'fisheye_expand_scale': 1.25,
    'fisheye_patch_scale': 1.5,
}
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 Tensor, ndim 3, dtype family float, shape (3, 8, 16) — shape (3, 8, 16) 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
genERP Ran
ran, honoured the contract we drafted
fanghua-yu/osrt/odisr/metrics/odi_metric.py
code served (permissive licence) · get_code("5753656385b79840")
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 math

def genERP(j, N):
    val = math.pi / N
    w = math.cos((j - N / 2 + 0.5) * val)
    return w

# The function takes two parameters:
# j: an index (integer)
# N: total number of samples (integer)

ARGS = {
    "j": 5,
    "N": 10
}
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 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
make_coord Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
fanghua-yu/osrt/odisr/utils/erp_downsample.py
code served (permissive licence) · get_code("0eb884d36fb9be8b")
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 (12, 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 torch

torch.manual_seed(0)

ARGS = {
    'shape': (4, 3),
    'ranges': [(-1, 1), (-1, 1)],
    'flatten': True,
    'double': False
}
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 (12, 2) — shape (12, 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
pers2erp Ran
ran on a synthesized input
fanghua-yu/osrt/odisr/utils/pers2erp.py
code served (permissive licence) · get_code("ecf6f60a59e10643")
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 int, shape (512, 1024, 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 cv2
import numpy as np

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

# Define arguments
ARGS = {
    "_img": _img,
    "PHI": 45.0,
    "wFOV": 90,
    "THETA": 30.0,
    "erp_size": [512, 1024]
}
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 ndarray, ndim 3, dtype family int, shape (512, 1024, 3) — shape (512, 1024, 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
test_TLC Ran
ran on a synthesized input
fanghua-yu/osrt/odisr/models/rcan_model.py
code served (permissive licence) · get_code("77ed7cd07319a3e6")
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, 3, 8, 7) 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 as nn

torch.manual_seed(0)

# Define a minimal model that satisfies the function's usage
class SimpleModel(nn.Module):
    def __init__(self, scale=1):
        super().__init__()
        self.scale = scale
        self.conv = nn.Conv2d(3, 3, kernel_size=3, padding=1)
    
    def forward(self, x):
        # x shape: (b, c, h, w)
        out = self.conv(x)
        # Apply upscaling if scale > 1
        if self.scale > 1:
            out = torch.nn.functional.interpolate(out, scale_factor=self.scale, mode='nearest')
        return out

# Create test data
batch_size = 2
channels = 3
height = 8
width = 7

img_lq = torch.randn(batch_size, channels, height, width)
model = SimpleModel(scale=1)
tile = 4
tile_overlap = 1
scale = 1

ARGS = {
    'img_lq': img_lq,
    'model': model,
    'tile': tile,
    'tile_overlap': tile_overlap,
    'scale': scale
}
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, 3, 8, 7) — shape (2, 3, 8, 7) 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
u Ran
ran, honoured the contract we drafted
fanghua-yu/osrt/odisr/utils/erp_downsample.py
code served (permissive licence) · get_code("7e200e5d7630e13b")
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 torch
import numpy as np

torch.manual_seed(0)

# The function u(s, a) takes two parameters:
# s: a value (likely scalar or array-like) that gets abs() applied and compared
# a: a scalar value used in polynomial calculations

# Looking at the function body:
# - s is used with abs(), compared with scalars, and raised to powers
# - a is used in arithmetic operations
# Both appear to be numeric values (int or float)

s = 0.5  # A value in the range [0, 1] to test the first branch
a = 2.0  # A scalar coefficient

ARGS = {
    "s": s,
    "a": a
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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
window_partition Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
fanghua-yu/osrt/odisr/archs/osrt_arch.py
code served (permissive licence) · get_code("f9c4a5ff2a55d09f")
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 (12, 4, 4, 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)

# Construct a valid input tensor
# x should have shape (b, h, w, c)
# We need h and w to be divisible by window_size
b = 2
h = 8
w = 12
c = 4
window_size = 4

x = torch.randn(b, h, w, c)

ARGS = {
    "x": x,
    "window_size": window_size,
}
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 (12, 4, 4, 4) — shape (12, 4, 4, 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
window_reverse Ran
ran; our contract draft was wrong, not the code
fanghua-yu/osrt/odisr/archs/osrt_arch.py
code served (permissive licence) · get_code("e641230dff80d607")
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, 4, 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 torch

torch.manual_seed(0)

# Parameters for window_reverse
window_size = 2
h = 4
w = 4
c = 3

# Calculate num_windows needed
# From the function: b = num_windows*b / (h*w / window_size / window_size)
# So: num_windows*b = b * (h*w / window_size / window_size)
# num_windows = (h*w / window_size / window_size)
num_windows = (h * w) // (window_size * window_size)
b = 2

# windows shape should be (num_windows*b, window_size, window_size, c)
windows = torch.randn(num_windows * b, window_size, window_size, c)

ARGS = {
    "windows": windows,
    "window_size": window_size,
    "h": h,
    "w": w,
}
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, 4, 4, 3) — shape (2, 4, 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
worker Ran
ran; our contract draft was wrong, not the code
fanghua-yu/osrt/odisr/utils/extract_subimage.py
code served (permissive licence) · get_code("d6a27ce76783f4d2")
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 str. 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 cv2
import numpy as np
import tempfile
import os

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

# Create a small dummy image file
dummy_img_path = os.path.join(temp_dir, 'test_image_x2.png')
dummy_img = np.random.randint(0, 256, (128, 128, 3), dtype=np.uint8)
cv2.imwrite(dummy_img_path, dummy_img)

# Construct the opt dictionary with all required keys
opt = {
    'crop_size': 64,
    'step': 32,
    'thresh_size': 10,
    'save_folder': temp_dir,
    'compression_level': 9,
    'wh': (96, 96),  # input shape (width, height)
    'scale': 1.0
}

ARGS = {
    'path': dummy_img_path,
    'opt': opt
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
make_model Not yet run
unverified (harvested, no recorded run)
fanghua-yu/osrt/odisr/archs/launet_arch.py
code served (permissive licence) · get_code("12ec7fbd4ce6cf35")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 9 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 1 the same month, 8 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.

FunctionAlso present inEarliest-datedOthers named
drop_path 224 other papers
71 dated earlier, 3 the same month, 143 later, 6 same year, order not determinable, 1 undated · 38 only as an assembly brick
2003.12039 (2020-03)
RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
2007.09384 (2020-07)
Multi-Scale Positive Sample Refinement for Few-Shot Object Detection
2010.11929 (2020-10)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
and 221 more not named here
window_partition 12 other papers
2 dated earlier, 9 later, 1 same year, order not determinable · 1 only as an assembly brick
2108.10257 (2021-08)
SwinIR: Image Restoration Using Swin Transformer
2203.11589 (2022-03)
Adaptive Patch Exiting for Scalable Single Image Super-Resolution
and 9 more not named here
window_reverse 11 other papers
2 dated earlier, 8 later, 1 same year, order not determinable · 1 only as an assembly brick
2108.10257 (2021-08)
SwinIR: Image Restoration Using Swin Transformer
2203.11589 (2022-03)
Adaptive Patch Exiting for Scalable Single Image Super-Resolution
and 8 more not named here
default_conv 1 other paper
1 dated earlier
Deng_LAU-Net_Latitude_Adaptive_Upscaling_Network_for_Omnidirectional_Image_Super-Resolution_CVPR_2021_paper (2021 (year only))
LAU-Net: Latitude Adaptive Upscaling Network for Omnidirectional Image Super-resolution
erp_downsample_fisheye 1 other paper
1 later
2402.19289 (2024-02)
CAMixerSR: Only Details Need More "Attention"
genERP 1 other paper
1 later
2402.19289 (2024-02)
CAMixerSR: Only Details Need More "Attention"
make_coord 1 other paper
1 later
2402.19289 (2024-02)
CAMixerSR: Only Details Need More "Attention"
u 1 other paper
1 later
2402.19289 (2024-02)
CAMixerSR: Only Details Need More "Attention"
worker 1 other paper
1 later
2402.19289 (2024-02)
CAMixerSR: Only Details Need More "Attention"

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

Omnidirectional images (ODIs) have obtained lots of research interest for immersive experiences. Although ODIs require extremely high resolution to capture details of the entire scene, the resolutions of most ODIs are insufficient. Previous methods attempt to solve this issue by image super-resolution (SR) on equirectangular projection (ERP) images. However, they omit geometric properties of ERP in the degradation process, and their models can hardly generalize to real ERP images. In this paper, we propose Fisheye downsampling, which mimics the real-world imaging process and synthesizes more realistic low-resolution samples. Then we design a distortion-aware Transformer (OSRT) to modulate ERP distortions continuously and self-adaptively. Without a cumbersome process, OSRT outperforms previous methods by about 0.2dB on PSNR. Moreover, we propose a convenient data augmentation strategy, which synthesizes pseudo ERP images from plain images. This simple strategy can alleviate the over-fitting problem of large networks and significantly boost the performance of ODISR. Extensive experiments have demonstrated the state-of-theart performance of our OSRT.

For agents

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

get_harvested_code_for_paper("2302.03453")
get_code_for_paper("2302.03453")
have("2302.03453")

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

curl https://syntology.ai/api/ran/2302.03453.json

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

[![Syntology run record](https://syntology.ai/api/ran/2302.03453.svg)](https://syntology.ai/paper/2302.03453)

Connect an agent — have() is free.