SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.13987 · ECCV · 2024

RealViformer: Investigating Attention for Real-World Video Super-Resolution

Angela Yao, Yuehan Zhang

arXiv · PDF · Open in the Atlas

Code that ran

19 of 23 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 9 have no contract verdict; 10 failures of our instrument; 4 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/2407.13987.json · badge: /api/ran/2407.13987.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
yuehan717/realviformer canonical (classifier label)
classifier basis: census
9 ran · 0 no recorded run
Yuehan717/RealViformer community
Papers with Code archive
10 ran · 4 no recorded run
FunctionStatusWhere it lives
Attention Ran
ran on a synthesized input
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("c673f82c27d694b0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, dtype family float, shape (2, 4, 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 torch
from torch import nn as nn
from torch.nn import functional as F
from einops import rearrange

class Attention(nn.Module):

    def __init__(self, dim, num_heads, bias, withmask=False):
        super(Attention, self).__init__()
        self.num_heads = num_heads
        self.temperature = nn.Parameter(torch.ones(num_heads, 1, 1))
        self.qkv = nn.Conv2d(dim, dim * 3, kernel_size=1, bias=bias)
        self.qkv_dwconv = nn.Conv2d(dim * 3, dim * 3, kernel_size=3, stride=1, padding=1, groups=dim * 3, bias=bias)
        self.project_out = nn.Conv2d(dim, dim, kernel_size=1, bias=bias)
        self.withmask = withmask
        if withmask:
            self.max_dsc = nn.AdaptiveMaxPool1d(1)
            self.avg_dsc = nn.AdaptiveAvgPool1d(1)
            self.linear1 = nn.Linear(2, 1, bias=bias)
            self.linear2 = nn.Linear(dim // num_heads, dim // num_heads, bias=bias)

    def forward(self, x):
        b, c, h, w = x.shape
        qkv = self.qkv_dwconv(self.qkv(x))
        q, k, v = qkv.chunk(3, dim=1)
        q = rearrange(q, 'b (head c) h w -> b head c (h w)', head=self.num_heads)
        k = rearrange(k, 'b (head c) h w -> b head c (h w)', head=self.num_heads)
        v = rearrange(v, 'b (head c) h w -> b head c (h w)', head=self.num_heads)
        q = torch.nn.functional.normalize(q, dim=-1)
        k = torch.nn.functional.normalize(k, dim=-1)
        attn = q @ k.transpose(-2, -1) * self.temperature
        if self.withmask:
            mask_in = attn.clone()
            b_, h_, c_, _ = mask_in.shape
            dsc = torch.cat([self.max_dsc(mask_in.reshape(b_, h_ * c_, c_)), self.avg_dsc(mask_in.reshape((b_, h_ * c_, c_)))], dim=-1).reshape(b_, h_, c_, 2)
            mask = self.linear1(dsc)
            mask = self.linear2(F.gelu(mask).transpose(-2, -1)).transpose(-2, -1)
            mask = F.sigmoid(mask)
        attn = attn.softmax(dim=-1)
        out = attn @ v
        if self.withmask:
            out = out * mask
        out = rearrange(out, 'b head c (h w) -> b (head c) h w', head=self.num_heads, h=h, w=w)
        out = self.project_out(out)
        return out


INIT_ARGS = {
    'dim': 4,
    'num_heads': 2,
    'bias': True,
    'withmask': True,
}

ARGS = {
    'x': torch.randn(2, 4, 3, 4),
}

if __name__ == '__main__':
    model = Attention(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Output: {output}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
einops, 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, 3, 4) — shape (2, 4, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
BasicModule Ran
ran on a synthesized input
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("34544c8f80a989ea")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, dtype family float, shape (2, 2, 16, 20) 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
from torch import nn as nn

class BasicModule(nn.Module):
    """Basic Module for SpyNet.
    """

    def __init__(self):
        super(BasicModule, self).__init__()
        self.basic_module = nn.Sequential(nn.Conv2d(in_channels=8, out_channels=32, kernel_size=7, stride=1, padding=3), nn.ReLU(inplace=False), nn.Conv2d(in_channels=32, out_channels=64, kernel_size=7, stride=1, padding=3), nn.ReLU(inplace=False), nn.Conv2d(in_channels=64, out_channels=32, kernel_size=7, stride=1, padding=3), nn.ReLU(inplace=False), nn.Conv2d(in_channels=32, out_channels=16, kernel_size=7, stride=1, padding=3), nn.ReLU(inplace=False), nn.Conv2d(in_channels=16, out_channels=2, kernel_size=7, stride=1, padding=3))

    def forward(self, tensor_input):
        return self.basic_module(tensor_input)


INIT_ARGS = {}

ARGS = {
    'tensor_input': torch.randn(2, 8, 16, 20)
}

if __name__ == '__main__':
    module = BasicModule(**INIT_ARGS)
    output = module(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Expected shape: torch.Size([2, 2, 16, 20])")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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, 2, 16, 20) — shape (2, 2, 16, 20) 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
BiasFree_LayerNorm Ran
ran on a synthesized input
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("7e2a41bb01b94fc9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) 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
from torch import nn as nn
import numbers

class BiasFree_LayerNorm(nn.Module):

    def __init__(self, normalized_shape):
        super(BiasFree_LayerNorm, self).__init__()
        if isinstance(normalized_shape, numbers.Integral):
            normalized_shape = (normalized_shape,)
        normalized_shape = torch.Size(normalized_shape)
        assert len(normalized_shape) == 1
        self.weight = nn.Parameter(torch.ones(normalized_shape))
        self.normalized_shape = normalized_shape

    def forward(self, x):
        sigma = x.var(-1, keepdim=True, unbiased=False)
        return x / torch.sqrt(sigma + 1e-05) * self.weight


INIT_ARGS = {
    'normalized_shape': 4
}

ARGS = {
    'x': torch.randn(2, 4)
}

if __name__ == '__main__':
    model = BiasFree_LayerNorm(**INIT_ARGS)
    output = model(**ARGS)
    print(output)
    print(output.shape)
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
CrossChannelAttention Ran
ran on a synthesized input
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("f3e998138ec90085")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, dtype family float, shape (2, 4, 3, 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
from torch import nn as nn
from einops import rearrange

class CrossChannelAttention(nn.Module):

    def __init__(self, dim_q, dim_k, num_heads, bias):
        super(CrossChannelAttention, self).__init__()
        self.num_heads = num_heads
        self.temperature = nn.Parameter(torch.ones(num_heads, 1, 1))
        self.kv = nn.Conv2d(dim_k, dim_k * 2, kernel_size=1, bias=bias)
        self.kv_dwconv = nn.Conv2d(dim_k * 2, dim_k * 2, kernel_size=3, stride=1, padding=1, groups=dim_k * 2, bias=bias)
        self.q = nn.Conv2d(dim_q, dim_q, kernel_size=3, stride=1, padding=1, bias=bias)
        self.project_out = nn.Conv2d(dim_q, dim_q, kernel_size=1, bias=bias)

    def forward(self, x, y, return_attn=False):
        b, c_kv, h, w = x.shape
        kv = self.kv_dwconv(self.kv(y))
        k, v = kv.chunk(2, dim=1)
        q = self.q(x)
        q = rearrange(q, 'b (head c) h w -> b head c (h w)', head=self.num_heads)
        k = rearrange(k, 'b (head c) h w -> b head c (h w)', head=self.num_heads)
        v = rearrange(v, 'b (head c) h w -> b head c (h w)', head=self.num_heads)
        q = torch.nn.functional.normalize(q, dim=-1)
        k = torch.nn.functional.normalize(k, dim=-1)
        attn = q @ k.transpose(-2, -1) * self.temperature
        attn = attn.softmax(dim=-1)
        out = attn @ v
        out = rearrange(out, 'b head c (h w) -> b (head c) h w', head=self.num_heads, h=h, w=w)
        out = self.project_out(out)
        if return_attn:
            return (out, attn)
        else:
            return out


INIT_ARGS = {
    'dim_q': 4,
    'dim_k': 4,
    'num_heads': 2,
    'bias': True,
}

ARGS = {
    'x': torch.randn(2, 4, 3, 3),
    'y': torch.randn(2, 4, 3, 3),
    'return_attn': False,
}

if __name__ == '__main__':
    model = CrossChannelAttention(**INIT_ARGS)
    output = model(**ARGS)
    print("Output shape:", output.shape)
    print("Success!")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
einops, 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, 3, 3) — shape (2, 4, 3, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
FeedForward Ran
ran on a synthesized input
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("1e98c6126e991a81")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, dtype family float, shape (2, 4, 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
from torch import nn as nn
from torch.nn import functional as F

class FeedForward(nn.Module):

    def __init__(self, dim, ffn_expansion_factor, bias):
        super(FeedForward, self).__init__()
        hidden_features = int(dim * ffn_expansion_factor)
        self.project_in = nn.Conv2d(dim, hidden_features * 2, kernel_size=1, bias=bias)
        self.dwconv = nn.Conv2d(hidden_features * 2, hidden_features * 2, kernel_size=3, stride=1, padding=1, groups=hidden_features * 2, bias=bias)
        self.project_out = nn.Conv2d(hidden_features, dim, kernel_size=1, bias=bias)

    def forward(self, x):
        x = self.project_in(x)
        x1, x2 = self.dwconv(x).chunk(2, dim=1)
        x = F.gelu(x1) * x2
        x = self.project_out(x)
        return x

INIT_ARGS = {
    'dim': 4,
    'ffn_expansion_factor': 2,
    'bias': True
}

ARGS = {
    'x': torch.randn(2, 4, 3, 5)
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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, 3, 5) — shape (2, 4, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
LayerNorm Ran
ran on a synthesized input
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("f65e81cd69bbf677")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, dtype family float, shape (2, 4, 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
from torch import nn as nn
from einops import rearrange
import numbers

class BiasFree_LayerNorm(nn.Module):

    def __init__(self, normalized_shape):
        super(BiasFree_LayerNorm, self).__init__()
        if isinstance(normalized_shape, numbers.Integral):
            normalized_shape = (normalized_shape,)
        normalized_shape = torch.Size(normalized_shape)
        assert len(normalized_shape) == 1
        self.weight = nn.Parameter(torch.ones(normalized_shape))
        self.normalized_shape = normalized_shape

    def forward(self, x):
        sigma = x.var(-1, keepdim=True, unbiased=False)
        return x / torch.sqrt(sigma + 1e-05) * self.weight

class WithBias_LayerNorm(nn.Module):

    def __init__(self, normalized_shape):
        super(WithBias_LayerNorm, self).__init__()
        if isinstance(normalized_shape, numbers.Integral):
            normalized_shape = (normalized_shape,)
        normalized_shape = torch.Size(normalized_shape)
        assert len(normalized_shape) == 1
        self.weight = nn.Parameter(torch.ones(normalized_shape))
        self.bias = nn.Parameter(torch.zeros(normalized_shape))
        self.normalized_shape = normalized_shape

    def forward(self, x):
        mu = x.mean(-1, keepdim=True)
        sigma = x.var(-1, keepdim=True, unbiased=False)
        return (x - mu) / torch.sqrt(sigma + 1e-05) * self.weight + self.bias

def to_3d(x):
    return rearrange(x, 'b c h w -> b (h w) c')

def to_4d(x, h, w):
    return rearrange(x, 'b (h w) c -> b c h w', h=h, w=w)

class LayerNorm(nn.Module):

    def __init__(self, dim, LayerNorm_type):
        super(LayerNorm, self).__init__()
        if LayerNorm_type == 'BiasFree':
            self.body = BiasFree_LayerNorm(dim)
        else:
            self.body = WithBias_LayerNorm(dim)

    def forward(self, x):
        h, w = x.shape[-2:]
        return to_4d(self.body(to_3d(x)), h, w)


INIT_ARGS = {
    'dim': 4,
    'LayerNorm_type': 'WithBias'
}

ARGS = {
    'x': torch.randn(2, 4, 3, 5)
}

if __name__ == '__main__':
    model = LayerNorm(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Output: {output}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
einops, 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, 3, 5) — shape (2, 4, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
SpyNet Ran
ran on a synthesized input
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("1098bd700b316e9f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, 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 torch
import math

# INIT_ARGS for SpyNet.__init__
INIT_ARGS = {
    'load_path': None  # No pretrained weights
}

# ARGS for SpyNet.forward
# Create two small RGB images (reference and support frames)
batch_size = 2
channels = 3
height = 64  # Must be divisible by 32 for the network
width = 96   # Different from height, also divisible by 32

ref_image = torch.randn(batch_size, channels, height, width)
supp_image = torch.randn(batch_size, channels, height, width)

ARGS = {
    'ref': ref_image,
    'supp': supp_image
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
Upsample Ran
ran on a synthesized input
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("dc176fc0afbcaf34")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, dtype family float, shape (2, 2, 6, 10) 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
from torch import nn

class Upsample(nn.Module):

    def __init__(self, n_feat):
        super(Upsample, self).__init__()
        self.body = nn.Sequential(nn.Conv2d(n_feat, n_feat * 2, kernel_size=3, stride=1, padding=1, bias=False), nn.PixelShuffle(2))

    def forward(self, x):
        return self.body(x)


INIT_ARGS = {
    'n_feat': 4
}

ARGS = {
    'x': torch.randn(2, 4, 3, 5)
}

if __name__ == '__main__':
    model = Upsample(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Input shape: {ARGS['x'].shape}")
    print(f"Output shape: {output.shape}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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, 2, 6, 10) — shape (2, 2, 6, 10) 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
WithBias_LayerNorm Ran
ran on a synthesized input
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("a19b3d60ae1aaffa")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) 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
from torch import nn as nn
import numbers

class WithBias_LayerNorm(nn.Module):

    def __init__(self, normalized_shape):
        super(WithBias_LayerNorm, self).__init__()
        if isinstance(normalized_shape, numbers.Integral):
            normalized_shape = (normalized_shape,)
        normalized_shape = torch.Size(normalized_shape)
        assert len(normalized_shape) == 1
        self.weight = nn.Parameter(torch.ones(normalized_shape))
        self.bias = nn.Parameter(torch.zeros(normalized_shape))
        self.normalized_shape = normalized_shape

    def forward(self, x):
        mu = x.mean(-1, keepdim=True)
        sigma = x.var(-1, keepdim=True, unbiased=False)
        return (x - mu) / torch.sqrt(sigma + 1e-05) * self.weight + self.bias


INIT_ARGS = {
    'normalized_shape': 4
}

ARGS = {
    'x': torch.randn(2, 4)
}

if __name__ == '__main__':
    model = WithBias_LayerNorm(**INIT_ARGS)
    output = model(**ARGS)
    print("Output shape:", output.shape)
    print("Output:", output)
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
flow_warp Ran
ran; our contract draft was wrong, not the code
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("ef9faf68f492a375")
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) and returned Tensor, ndim 4, dtype family float, shape (2, 4, 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 a small image/feature map tensor: (batch=2, channels=4, height=3, width=5)
x = torch.randn(2, 4, 3, 5)

# Create optical flow tensor: (batch=2, height=3, width=5, 2)
# The flow represents displacement in (x, y) directions
flow = torch.randn(2, 3, 5, 2) * 0.5  # Small displacements

ARGS = {
    'x': x,
    'flow': flow,
    'interp_mode': 'bilinear',
    'padding_mode': 'zeros',
    'align_corners': True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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, 3, 5) — shape (2, 4, 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
img2tensor Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yuehan717/realviformer/img_util.py
code served (permissive licence) · get_code("8b5cf3d00508f17b")
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, 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 cv2
import torch
import numpy as np

torch.manual_seed(0)

# Create a small BGR image as numpy array (height=4, width=3, channels=3)
img_bgr = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)

# Create a list of images for testing the list case
imgs_list = [img_bgr, img_bgr.copy()]

# Create single image for testing the single image case
imgs_single = img_bgr

ARGS = {
    "imgs": imgs_single,
    "bgr2rgb": True,
    "float32": True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy, torch — 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, 4, 3) — shape (3, 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
make_layer Ran
ran; our contract draft was wrong, not the code
yuehan717/realviformer/archs/arch_util.py
code served (permissive licence) · get_code("96ad5dc9ca239aec")
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) and returned Sequential. 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)

# Define a simple basic block that can be instantiated with **kwarg
class SimpleBasicBlock(nn.Module):
    def __init__(self, in_channels=3, out_channels=4):
        super().__init__()
        self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=3, padding=1)
    
    def forward(self, x):
        return self.conv(x)

# Construct the arguments
ARGS = {
    'basic_block': SimpleBasicBlock,
    'num_basic_block': 3,
    'in_channels': 3,
    'out_channels': 4
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
mod_crop Ran
ran; our contract draft was wrong, not the code
yuehan717/realviformer/data_util.py
code served (permissive licence) · get_code("223d0788b0725dfb")
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 (12, 16, 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 test image (2D grayscale)
img_2d = np.random.randint(0, 256, size=(10, 15), dtype=np.uint8)

# Create a small test image (3D RGB)
img_3d = np.random.randint(0, 256, size=(12, 18, 3), dtype=np.uint8)

scale = 4

ARGS = {
    "img": img_3d,
    "scale": scale,
}
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 3, dtype family int, shape (12, 16, 3) — shape (12, 16, 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
resize_flow Ran
ran; our contract draft was wrong, not the code
yuehan717/realviformer/archs/arch_util.py
code served (permissive licence) · get_code("9f330f530563e78f")
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, 2, 2, 6) 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 a small flow tensor: [N=2, C=2, H=4, W=3]
# Flow has 2 channels (for x and y components)
flow = torch.randn(2, 2, 4, 3)

# Test with 'ratio' size type
size_type = 'ratio'
sizes = [0.5, 2.0]  # ratio_h=0.5, ratio_w=2.0
interp_mode = 'bilinear'
align_corners = False

ARGS = {
    'flow': flow,
    'size_type': size_type,
    'sizes': sizes,
    'interp_mode': interp_mode,
    'align_corners': align_corners
}
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, 2, 2, 6) — shape (2, 2, 2, 6) 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
scandir Ran
ran; our contract draft was wrong, not the code
yuehan717/realviformer/data_util.py
code served (permissive licence) · get_code("f4e0ddb8e63a63ee")
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 generator. 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 os
import tempfile
import shutil

# Create a temporary directory structure for testing
temp_dir = tempfile.mkdtemp()

# Create some test files
os.makedirs(os.path.join(temp_dir, "subdir"), exist_ok=True)
with open(os.path.join(temp_dir, "file1.txt"), "w") as f:
    f.write("test")
with open(os.path.join(temp_dir, "file2.py"), "w") as f:
    f.write("test")
with open(os.path.join(temp_dir, "subdir", "file3.txt"), "w") as f:
    f.write("test")
with open(os.path.join(temp_dir, ".hidden"), "w") as f:
    f.write("test")

ARGS = {
    "dir_path": temp_dir,
    "suffix": ".txt",
    "recursive": True,
    "full_path": False
}
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 generator — a generator output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a generator 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
tensor2img Ran
ran; our contract draft was wrong, not the code
yuehan717/realviformer/img_util.py
code served (permissive licence) · get_code("6d6227e782335df3")
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 (14, 9, 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
import numpy as np

torch.manual_seed(0)

# Create a 4D tensor (B x C x H x W) - batch of 2, 3 channels, 4x5 spatial dims
tensor_4d = torch.rand(2, 3, 4, 5)

# Create a 3D tensor (C x H x W) - 3 channels, 6x7 spatial dims
tensor_3d = torch.rand(3, 6, 7)

# Create a 2D tensor (H x W) - 8x9 spatial dims
tensor_2d = torch.rand(8, 9)

# Create a list of tensors
tensor_list = [tensor_4d, tensor_3d]

ARGS = {
    'tensor': tensor_4d,
    'rgb2bgr': True,
    'out_type': np.uint8,
    'min_max': (0, 1)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy, torch, torchvision — 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 (14, 9, 3) — shape (14, 9, 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
tensor2img_fast Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yuehan717/realviformer/img_util.py
code served (permissive licence) · get_code("442bf3d188b77bf1")
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 (4, 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 torch
import cv2

torch.manual_seed(0)

# Create a tiny tensor with shape (1, c, h, w)
# Using distinct dimensions: batch=1, channels=3, height=4, width=5
tensor = torch.rand(1, 3, 4, 5)

# Clamp to valid range for image conversion
tensor = tensor.clamp(0, 1)

ARGS = {
    "tensor": tensor,
    "rgb2bgr": True,
    "min_max": (0, 1)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, torch — 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 (4, 5, 3) — shape (4, 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
to_3d Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yuehan717/realviformer/archs/realviformer_arch.py
pointer only (licence: MIT); the stored copy get_code serves, from gladzhang/Xformer, is pointer only (licence: NONE) · get_code("82a15cc1e46f7e4d")
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 gladzhang/Xformer, 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.

to_4d Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yuehan717/realviformer/archs/realviformer_arch.py
pointer only (licence: MIT); the stored copy get_code serves, from gladzhang/Xformer, is pointer only (licence: NONE) · get_code("b20f2a5df739a59e")
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 gladzhang/Xformer, 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.

ChannelAttentionBlock Not yet run
unverified (harvested, no recorded run)
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("181873d6f2194b59")
How it ran

no invocation recorded.

Downsample Not yet run
unverified (harvested, no recorded run)
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("4bde66e2e3722ca8")
How it ran

no invocation recorded.

RealViformer Not yet run
unverified (harvested, no recorded run)
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("8378fd449c8fc540")
How it ran

no invocation recorded.

TransformerBlock Not yet run
unverified (harvested, no recorded run)
Yuehan717/RealViformer/archs/realviformer_arch.py
code served (permissive licence) · get_code("e91c5e9a654fd7dc")
How it ran

no invocation recorded.

Also present in other papers' code

Of 23 samples lifted from this paper's linked repositories, 23 carry a function identity and 0 do not. Of the 23 distinct functions with an identity, 15 are also present in other papers whose code we hold (15 with at least one paper dated earlier, 6 the same month, 14 later) and 8 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
to_3d
brick of this paper's assembly: helper
80 other papers
49 dated earlier, 3 the same month, 26 later, 2 same year, order not determinable · 18 only as an assembly brick
2003.06792 (2020-03)
Learning Enriched Features for Real Image Restoration and Enhancement
2102.02808 (2021-02)
Multi-Stage Progressive Image Restoration
2111.09881 (2021-11)
Restormer: Efficient Transformer for High-Resolution Image Restoration
and 77 more not named here
to_4d
brick of this paper's assembly: helper
79 other papers
48 dated earlier, 3 the same month, 26 later, 2 same year, order not determinable · 18 only as an assembly brick
2003.06792 (2020-03)
Learning Enriched Features for Real Image Restoration and Enhancement
2102.02808 (2021-02)
Multi-Stage Progressive Image Restoration
2111.09881 (2021-11)
Restormer: Efficient Transformer for High-Resolution Image Restoration
and 76 more not named here
make_layer 29 other papers
18 dated earlier, 10 later, 1 same year, order not determinable
2107.10833 (2021-07)
Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data
2108.11048 (2021-08)
Memory-Augmented Non-Local Attention for Video Super-Resolution
and 26 more not named here
flow_warp
brick of this paper's assembly: helper
25 other papers
15 dated earlier, 9 later, 1 same year, order not determinable · 1 only as an assembly brick
2107.10833 (2021-07)
Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data
2303.06373 (2023-03)
Recursive Generalization Transformer for Image Super-Resolution
and 22 more not named here
BiasFree_LayerNorm
brick of this paper's assembly: helper
24 other papers
13 dated earlier, 1 the same month, 10 later · 20 only as an assembly brick
2102.02808 (2021-02)
Multi-Stage Progressive Image Restoration
2207.10040 (2022-07)
Single Frame Atmospheric Turbulence Mitigation: A Benchmark Study and A New Physics-Inspired Transformer Model
2303.06440 (2023-03)
Xformer: Hybrid X-Shaped Transformer for Image Denoising
and 21 more not named here
WithBias_LayerNorm
brick of this paper's assembly: helper
24 other papers
13 dated earlier, 1 the same month, 10 later · 20 only as an assembly brick
2102.02808 (2021-02)
Multi-Stage Progressive Image Restoration
2207.10040 (2022-07)
Single Frame Atmospheric Turbulence Mitigation: A Benchmark Study and A New Physics-Inspired Transformer Model
2303.06440 (2023-03)
Xformer: Hybrid X-Shaped Transformer for Image Denoising
and 21 more not named here
LayerNorm
brick of this paper's assembly: helper
22 other papers
13 dated earlier, 1 the same month, 8 later · 17 only as an assembly brick
2102.02808 (2021-02)
Multi-Stage Progressive Image Restoration
2207.10040 (2022-07)
Single Frame Atmospheric Turbulence Mitigation: A Benchmark Study and A New Physics-Inspired Transformer Model
2303.06440 (2023-03)
Xformer: Hybrid X-Shaped Transformer for Image Denoising
and 19 more not named here
Downsample
brick of this paper's assembly: model
18 other papers
9 dated earlier, 9 later · 16 only as an assembly brick
2102.02808 (2021-02)
Multi-Stage Progressive Image Restoration
2207.10040 (2022-07)
Single Frame Atmospheric Turbulence Mitigation: A Benchmark Study and A New Physics-Inspired Transformer Model
2303.06440 (2023-03)
Xformer: Hybrid X-Shaped Transformer for Image Denoising
and 15 more not named here
Upsample
brick of this paper's assembly: model
17 other papers
9 dated earlier, 8 later · 15 only as an assembly brick
2102.02808 (2021-02)
Multi-Stage Progressive Image Restoration
2207.10040 (2022-07)
Single Frame Atmospheric Turbulence Mitigation: A Benchmark Study and A New Physics-Inspired Transformer Model
2303.06440 (2023-03)
Xformer: Hybrid X-Shaped Transformer for Image Denoising
and 14 more not named here
FeedForward
brick of this paper's assembly: helper
13 other papers
7 dated earlier, 1 the same month, 5 later · 12 only as an assembly brick
2102.02808 (2021-02)
Multi-Stage Progressive Image Restoration
2207.10040 (2022-07)
Single Frame Atmospheric Turbulence Mitigation: A Benchmark Study and A New Physics-Inspired Transformer Model
2303.06440 (2023-03)
Xformer: Hybrid X-Shaped Transformer for Image Denoising
and 10 more not named here
BasicModule
brick of this paper's assembly: helper
3 other papers
2 dated earlier, 1 later · 2 only as an assembly brick
2204.07114 (2022-04)
Look Back and Forth: Video Super-Resolution with Explicit Temporal Difference Modeling
2406.13457 (2024-06)
EvTexture: Event-driven Texture Enhancement for Video Super-Resolution
2601.00533 (2026-01)
All-in-One Video Restoration under Smoothly Evolving Unknown Weather Degradations
img2tensor 3 other papers
1 dated earlier, 2 later
2309.03020 (2023-09)
SEAL: A Framework for Systematic Evaluation of Real-World Super-Resolution
2408.13239 (2024-08)
CustomCrafter: Customized Video Generation with Preserving Motion and Concept Composition Abilities
2410.00418 (2024-10)
Posterior-Mean Rectified Flow: Towards Minimum MSE Photo-Realistic Image Restoration
mod_crop 3 other papers
1 dated earlier, 2 later
2305.12966 (2023-05)
Hierarchical Integration Diffusion Model for Realistic Image Deblurring
2411.13383 (2024-11)
Adversarial Diffusion Compression for Real-World Image Super-Resolution
tensor2img 3 other papers
2 dated earlier, 1 later
2309.03020 (2023-09)
SEAL: A Framework for Systematic Evaluation of Real-World Super-Resolution
2401.05907 (2024-01)
Efficient Image Deblurring Networks based on Diffusion Models
2412.01427 (2024-12)
FoundIR: Unleashing Million-scale Training Data to Advance Foundation Models for Image Restoration
tensor2img_fast 1 other paper
1 dated earlier
2309.03020 (2023-09)
SEAL: A Framework for Systematic Evaluation of Real-World Super-Resolution

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 real-world video super-resolution (VSR), videos suffer from in-the-wild degradations and artifacts. VSR methods, especially recurrent ones, tend to propagate artifacts over time in the real-world setting and are more vulnerable than image super-resolution. This paper investigates the influence of artifacts on commonly used covariance-based attention mechanisms in VSR. Comparing the widely-used spatial attention, which computes covariance over space, versus the channel attention, we observe that the latter is less sensitive to artifacts. However, channel attention leads to feature redundancy, as evidenced by the higher covariance among output channels. As such, we explore simple techniques such as the squeeze-excite mechanism and covariance-based rescaling to counter the effects of high channel covariance. Based on our findings, we propose RealViformer. This channel-attention-based realworld VSR framework surpasses state-of-the-art on two real-world VSR datasets with fewer parameters and faster runtimes. The source code is available at https://github.com/Yuehan717/RealViformer.

For agents

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

get_harvested_code_for_paper("2407.13987")
get_code_for_paper("2407.13987")
have("2407.13987")

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

curl https://syntology.ai/api/ran/2407.13987.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.13987.svg)](https://syntology.ai/paper/2407.13987)

Connect an agent — have() is free.