SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2404.13153 · CVPR · 2024

Motion-adaptive Separable Collaborative Filters for Blind Motion Deblurring

Ming-Hsuan Yang, Shuai Li, Xueming Qian, Xiangyu Xu, Xuan Wang, Chengxu Liu, Ruhao Tian

arXiv · PDF · Open in the Atlas

Code that ran

10 of 13 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, 2 violated it and 0 have no contract verdict; 7 failures of our instrument; 3 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/2404.13153.json · badge: /api/ran/2404.13153.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
ChengxuLiu/MISCFilter canonical (classifier label)
classifier basis: census
10 ran · 3 no recorded run
FunctionStatusWhere it lives
CharbonnierFunc Ran
ran; our contract draft was wrong, not the code
ChengxuLiu/MISCFilter/models/MISCFilterNet.py
code served (permissive licence) · get_code("9e8e56135365289e")
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 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

torch.manual_seed(0)

# The function takes 'data' (a tensor) and an optional 'epsilon' parameter
# It computes the Charbonnier loss: mean(sqrt(data^2 + epsilon^2))
# So 'data' should be a tensor of numeric values

data = torch.randn(2, 4)
epsilon = 0.001

ARGS = {
    "data": data,
    "epsilon": epsilon,
}
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 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
compute_psnr Ran
ran, violated the contract we drafted
ChengxuLiu/MISCFilter/evaluate_RealBlur.py
pointer only (licence: MIT); the stored copy get_code serves, from invokerer/deeprft, is pointer only (licence: NOASSERTION) · get_code("91fa50e293bf2fde")
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 invokerer/deeprft, 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.

flow_warp Ran
ran; our contract draft was wrong, not the code
ChengxuLiu/MISCFilter/models/MISCFilterNet.py
code served (permissive licence) · get_code("dfd71b86aa96a884")
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, 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 input tensor: (batch=2, channels=4, height=3, width=5)
x = torch.randn(2, 4, 3, 5, dtype=torch.float32)

# Create a flow tensor with matching spatial dimensions: (batch=2, height=3, width=5, 2)
# The last dimension is 2 for (x, y) flow components
flow = torch.randn(2, 3, 5, 2, dtype=torch.float32)

ARGS = {
    'x': x,
    'flow': flow,
    'interpolation': 'bilinear',
    'padding_mode': 'zeros',
    'align_corners': 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 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
is_frozen Ran
ran, violated the contract we drafted
ChengxuLiu/MISCFilter/utils/model_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from invokerer/deeprft, is pointer only (licence: NOASSERTION) · get_code("930cddf0a4d80ab1")
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 invokerer/deeprft, 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.

load_start_epoch Ran
ran, honoured the contract we drafted
ChengxuLiu/MISCFilter/utils/model_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from invokerer/deeprft, is pointer only (licence: NOASSERTION) · get_code("c3bea8c4ee6bef4a")
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 invokerer/deeprft, 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.

moduleNormalize Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ChengxuLiu/MISCFilter/models/utility.py
code served (permissive licence) · get_code("9a2082456dfd29d9")
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, 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)

# The function expects a frame tensor with shape (batch, channels, height, width)
# where channels = 3 (RGB)
# It subtracts mean values from each channel independently
frame = torch.randn(2, 3, 4, 4)

ARGS = {
    "frame": frame
}
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, 4, 4) — shape (2, 3, 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
to_variable Ran
ran; our contract draft was wrong, not the code
ChengxuLiu/MISCFilter/models/utility.py
code served (permissive licence) · get_code("989c91edb537cd8f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, 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.autograd import Variable

torch.manual_seed(0)

# Create a small tensor to pass to to_variable
x = torch.randn(2, 4)

# Define ARGS mapping all parameters of to_variable
ARGS = {
    'x': x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
window_partitions Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ChengxuLiu/MISCFilter/models/layers.py
code served (permissive licence) · get_code("6246e43ffc3d3ad0")
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 (24, 4, 2, 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)

# The function expects:
# - x: a tensor of shape (B, C, H, W) where H and W are divisible by window_size
# - window_size: an integer that divides both H and W

# Create a small tensor with distinct dimensions for probing
# B=2, C=4, H=8, W=6 (all divisible by window_size=2)
B = 2
C = 4
H = 8
W = 6
window_size = 2

x = torch.randn(B, C, H, W)

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 (24, 4, 2, 2) — shape (24, 4, 2, 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
window_partitionx Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ChengxuLiu/MISCFilter/models/layers.py
code served (permissive licence) · get_code("3bbb99d8422b5bd6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create a small test tensor: (B=2, C=4, H=8, W=8)
# Using distinct sizes: B=2, C=4, H=8, W=8
x = torch.randn(2, 4, 8, 8)

# window_size must divide H and W evenly for the main partition
window_size = 4

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
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
window_reverses Ran
ran; our contract draft was wrong, not the code
ChengxuLiu/MISCFilter/models/layers.py
code served (permissive licence) · get_code("89c554708e3dae51")
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, 4, 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

torch.manual_seed(0)

# Parameters for window_reverses
window_size = 2
H = 4
W = 6
C = 3
B = 2

# Create windows tensor: (num_windows*B, C, window_size, window_size)
# num_windows = (H // window_size) * (W // window_size) = 2 * 3 = 6
# So windows shape should be (6*B, C, window_size, window_size) = (12, 3, 2, 2)
num_windows = (H // window_size) * (W // window_size)
windows = torch.randn(num_windows * B, C, window_size, window_size)

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, 3, 4, 6) — shape (2, 3, 4, 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
compute_ssim Not yet run
unverified (harvested, no recorded run)
ChengxuLiu/MISCFilter/evaluate_RealBlur.py
pointer only (licence: MIT); the stored copy get_code serves, from invokerer/deeprft, is pointer only (licence: NOASSERTION) · get_code("464c3d3b7a1eb4b3")
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 invokerer/deeprft, 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. No invocation recorded.

cupy_kernel Not yet run
unverified (harvested, no recorded run)
ChengxuLiu/MISCFilter/models/MISCKernel_cuda.py
code served (permissive licence) · get_code("50fe0d31f1836940")
How it ran

no invocation recorded.

image_align Not yet run
unverified (harvested, no recorded run)
ChengxuLiu/MISCFilter/evaluate_RealBlur.py
pointer only (licence: MIT); the stored copy get_code serves, from pp00704831/Stripformer-ECCV-2022-, is pointer only (licence: NOASSERTION) · get_code("5038ac50ef83af3a")
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 pp00704831/Stripformer-ECCV-2022-, 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. No invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 10 are also present in other papers whose code we hold (10 with at least one paper dated earlier, 1 the same month, 6 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
is_frozen 8 other papers
4 dated earlier, 3 later, 1 same year, order not determinable
Zhang_Learning_Rain_Location_Prior_for_Nighttime_Deraining_ICCV_2023_paper (2023 (year only))
Learning Rain Location Prior for Nighttime Deraining
2308.03867 (2023-08)
From Sky to the Ground: A Large-scale Benchmark and Simple Baseline Towards Real Rain Removal
2309.03063 (2023-09)
Prompt-based Ingredient-Oriented All-in-One Image Restoration
and 5 more not named here
load_start_epoch 8 other papers
4 dated earlier, 3 later, 1 same year, order not determinable
Zhang_Learning_Rain_Location_Prior_for_Nighttime_Deraining_ICCV_2023_paper (2023 (year only))
Learning Rain Location Prior for Nighttime Deraining
2308.03867 (2023-08)
From Sky to the Ground: A Large-scale Benchmark and Simple Baseline Towards Real Rain Removal
2309.03063 (2023-09)
Prompt-based Ingredient-Oriented All-in-One Image Restoration
and 5 more not named here
compute_psnr 5 other papers
3 dated earlier, 2 later
2201.02973 (2022-01)
MAXIM: Multi-Axis MLP for Image Processing
2305.12966 (2023-05)
Hierarchical Integration Diffusion Model for Realistic Image Deblurring
2401.00027 (2024-01)
Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring
and 2 more not named here
image_align 5 other papers
3 dated earlier, 2 later
2201.02973 (2022-01)
MAXIM: Multi-Axis MLP for Image Processing
2305.12966 (2023-05)
Hierarchical Integration Diffusion Model for Realistic Image Deblurring
2401.00027 (2024-01)
Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring
and 2 more not named here
compute_ssim 4 other papers
2 dated earlier, 2 later
2201.02973 (2022-01)
MAXIM: Multi-Axis MLP for Image Processing
2401.00027 (2024-01)
Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring
2405.16749 (2024-05)
DMPlug: A Plug-in Method for Solving Inverse Problems with Diffusion Models
and 1 more not named here
CharbonnierFunc 3 other papers
2 dated earlier, 1 same year, order not determinable
1907.10244 (2019-07)
AdaCoF: Adaptive Collaboration of Flows for Video Frame Interpolation
2303.09508 (2023-03)
LDMVFI: Video Frame Interpolation with Latent Diffusion Models
aaai_27912 (2024 (year only))
LDMVFI: Video Frame Interpolation with Latent Diffusion Models
window_partitionx 3 other papers
1 dated earlier, 1 the same month, 1 later
2401.08171 (2024-01)
Deep Linear Array Pushbroom Image Restoration: A Degradation Pipeline and Jitter-Aware Restoration Network
2404.01547 (2024-04)
Bidirectional Multi-Scale Implicit Neural Representations for Image Deraining
2406.09135 (2024-06)
AdaRevD: Adaptive Patch Exiting Reversible Decoder Pushes the Limit of Image Deblurring
cupy_kernel 2 other papers
1 dated earlier, 1 same year, order not determinable
2303.09508 (2023-03)
LDMVFI: Video Frame Interpolation with Latent Diffusion Models
aaai_27912 (2024 (year only))
LDMVFI: Video Frame Interpolation with Latent Diffusion Models
moduleNormalize 2 other papers
1 dated earlier, 1 same year, order not determinable
2303.09508 (2023-03)
LDMVFI: Video Frame Interpolation with Latent Diffusion Models
aaai_27912 (2024 (year only))
LDMVFI: Video Frame Interpolation with Latent Diffusion Models
to_variable 2 other papers
2 dated earlier
1905.13613 (2019-05)
Regression Networks for Meta-Learning Few-Shot Classification
1907.10244 (2019-07)
AdaCoF: Adaptive Collaboration of Flows for Video Frame Interpolation

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

Eliminating image blur produced by various kinds of motion has been a challenging problem. Dominant approaches rely heavily on model capacity to remove blurring by reconstructing residual from blurry observation in feature space. These practices not only prevent the capture of spatially variable motion in the real world but also ignore the tailored handling of various motions in image space. In this paper, we propose a novel real-world deblurring filtering model called the Motion-adaptive Separable Collaborative (MISC) Filter. In particular, we use a motion estimation network to capture motion information from neighborhoods, thereby adaptively estimating spatially-variant motion flow, mask, kernels, weights, and offsets to obtain the MISC Filter. The MISC Filter first aligns the motion-induced blurring patterns to the motion middle along the predicted flow direction, and then collaboratively filters the aligned image through the predicted kernels, weights, and offsets to generate the output. This design can handle more generalized and complex motion in a spatially differentiated manner. Furthermore, we analyze the relationships between the motion estimation network and the residual reconstruction network. Extensive experiments on four widely used benchmarks demonstrate that our method provides an effective solution for real-world motion blur removal and achieves state-of-the-art performance. Code is available at https: //github.com/ChengxuLiu/MISCFilter.

For agents

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

get_harvested_code_for_paper("2404.13153")
get_code_for_paper("2404.13153")
have("2404.13153")

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

curl https://syntology.ai/api/ran/2404.13153.json

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

[![Syntology run record](https://syntology.ai/api/ran/2404.13153.svg)](https://syntology.ai/paper/2404.13153)

Connect an agent — have() is free.