SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.11591 · 2023

SVCNet: Scribble-based Video Colorization Network with Temporal Aggregation

arXiv · PDF · Open in the Atlas

Code that ran

10 of 11 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 5 have no contract verdict; 5 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/2303.11591.json · badge: /api/ran/2303.11591.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
zhaoyuzhi/svcnet canonical (classifier label)
classifier basis not recorded on the link
10 ran · 1 no recorded run
FunctionStatusWhere it lives
PWCEstimate Ran
ran; our contract draft was wrong, not the code
zhaoyuzhi/svcnet/GCS/pwcnet.py
code served (permissive licence) · get_code("215ab102388c6606")
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, 32, 48) 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)

class MockPWCNet(nn.Module):
    """Minimal PWCNet mock that takes two tensors and outputs optical flow."""
    def __init__(self):
        super().__init__()
        self.conv = nn.Conv2d(6, 2, kernel_size=3, padding=1)
    
    def forward(self, tensorFirst, tensorSecond):
        # Concatenate inputs and output 2-channel flow map
        x = torch.cat([tensorFirst, tensorSecond], dim=1)
        flow = self.conv(x)
        return flow

# Create a mock PWCNet instance
pwcnet = MockPWCNet()

# Create tiny test tensors: batch=2, channels=3, height=32, width=48
# Using distinct sizes (32x48) to make dimensions probed later unambiguous
tensorFirst = torch.randn(2, 3, 32, 48)
tensorSecond = torch.randn(2, 3, 32, 48)

ARGS = {
    'pwcnet': pwcnet,
    'tensorFirst': tensorFirst,
    'tensorSecond': tensorSecond,
    'drange': False,
    'reshape': 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 4, dtype family float, shape (2, 2, 32, 48) — shape (2, 2, 32, 48) 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
PWCNetBackward Ran
ran on a synthesized input
zhaoyuzhi/svcnet/GCS/pwcnet.py
code served (permissive licence) · get_code("0fbed6e4eb096e61")
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, 10) 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
import torch.nn.functional as F

torch.manual_seed(0)

# Construct tensorInput: [batch, channels, height, width]
# The function expects this to have size(3) and size(2) for width and height normalization
tensorInput = torch.randn(2, 3, 8, 10)

# Construct tensorFlow: [batch, 2, height, width]
# Flow has 2 channels (horizontal and vertical flow)
# Must have same spatial dimensions as tensorInput for grid_sample
tensorFlow = torch.randn(2, 2, 8, 10)

ARGS = {
    'tensorInput': tensorInput,
    'tensorFlow': tensorFlow
}
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, 10) — shape (2, 3, 8, 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: none by the harness; any seed is in the setup block
Reshape_Tensor Ran
ran; our contract draft was wrong, not the code
zhaoyuzhi/svcnet/GCS/pwcnet.py
code served (permissive licence) · get_code("bf90a9d890668ca8")
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 torch.nn.functional as F
import math

torch.manual_seed(0)

# Create a small tensor with shape (B=2, C=4, H=3, W=5)
tensor = torch.randn(2, 4, 3, 5)

# Define basesize parameter
basesize = 64.0

# Construct ARGS dict with all parameters
ARGS = {
    'tensor': tensor,
    'basesize': basesize
}
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 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
color_scribble Ran
ran on a synthesized input
zhaoyuzhi/svcnet/CPNet/validation.py
code served (permissive licence) · get_code("8efa57724ffd7b8b")
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 (8, 6, 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 with shape (height, width, channels)
img = np.random.randint(0, 256, size=(8, 6, 3), dtype=np.uint8)

# Define parameters
color_point = 5  # Number of scribble points
color_width = 3  # Width of the scribble brush

ARGS = {
    "img": img,
    "color_point": color_point,
    "color_width": color_width,
}
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 (8, 6, 3) — shape (8, 6, 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
convert_lab_to_bgr Ran
ran on a synthesized input
zhaoyuzhi/svcnet/SSNet/validation.py
code served (permissive licence) · get_code("3a7d6eb7ccf9cd56")
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, 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
import numpy as np

torch.manual_seed(0)

# Create input_img: batch of 1, 1 channel (L channel), 4x3 spatial dimensions
input_img = torch.randn(1, 1, 4, 3)

# Create out_img: batch of 1, 2 channels (a and b channels), 4x3 spatial dimensions
out_img = torch.randn(1, 2, 4, 3)

ARGS = {
    'input_img': input_img,
    'out_img': out_img
}
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 ndarray, ndim 3, dtype family int, shape (4, 3, 3) — shape (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: none by the harness; any seed is in the setup block
get_confusion_matrix Ran
ran on a synthesized input
zhaoyuzhi/svcnet/Evaluation/mIoU_Evaluation/svcnet_compute_miou.py
code served (permissive licence) · get_code("2db313c7eabe0246")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (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 numpy as np

# Analyze the function to understand what arguments are needed:
# - label: array-like, used with [None] and indexing, compared with ignore value
# - pred: array-like, used with [None] and indexing
# - size: passed as parameter but never used in function body
# - num_class: int, used in range() and arithmetic
# - ignore: int, default -1, compared with values in label

# Create small test arrays
label = np.array([0, 1, 0, 1, 2], dtype='int32')
pred = np.array([0, 1, 1, 1, 2], dtype='int32')
size = 5  # arbitrary, not used
num_class = 3
ignore = -1  # default value, won't match any label values

ARGS = {
    'label': label,
    'pred': pred,
    'size': size,
    'num_class': num_class,
    'ignore': ignore
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (3, 3) — shape (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: none by the harness; any seed is in the setup block
get_files Ran
ran; our contract draft was wrong, not the code
zhaoyuzhi/svcnet/GCS/generate_color_scribbles_ImageNet.py
code served (permissive licence) · get_code("889fe30d9fa3b00d")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import tempfile
import shutil

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

# Create some subdirectories and files
subdir1 = os.path.join(temp_dir, "subdir1")
subdir2 = os.path.join(temp_dir, "subdir2")
os.makedirs(subdir1)
os.makedirs(subdir2)

# Create some test files
open(os.path.join(temp_dir, "file1.txt"), "w").close()
open(os.path.join(temp_dir, "file2.txt"), "w").close()
open(os.path.join(subdir1, "file3.txt"), "w").close()
open(os.path.join(subdir2, "file4.txt"), "w").close()

ARGS = {
    "path": temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_jpgs Ran
ran; our contract draft was wrong, not the code
zhaoyuzhi/svcnet/GCS/generate_color_scribbles_ImageNet.py
code served (permissive licence) · get_code("a7688ed9fd9208bc")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import tempfile
import shutil

# Create a temporary directory structure with some files
temp_dir = tempfile.mkdtemp()

# Create subdirectories and files
os.makedirs(os.path.join(temp_dir, "subdir1"), exist_ok=True)
os.makedirs(os.path.join(temp_dir, "subdir2"), exist_ok=True)

# Create some dummy files
open(os.path.join(temp_dir, "file1.jpg"), 'w').close()
open(os.path.join(temp_dir, "file2.txt"), 'w').close()
open(os.path.join(temp_dir, "subdir1", "file3.jpg"), 'w').close()
open(os.path.join(temp_dir, "subdir2", "file4.png"), 'w').close()

ARGS = {
    "path": temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
recover_ndarray_to_tensor Ran
ran on a synthesized input
zhaoyuzhi/svcnet/CPNet/validation.py
code served (permissive licence) · get_code("701ea0de7c87d405")
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 (1, 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 numpy as np
import torch

torch.manual_seed(0)

# Create a small test image as numpy array
# Shape: (height, width, channels) = (4, 3, 3) for RGB image
img = np.random.randint(0, 256, size=(4, 3, 3), dtype=np.uint8)

ARGS = {
    "img": img
}
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 Tensor, ndim 4, dtype family float, shape (1, 3, 4, 3) — shape (1, 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
text_readlines Ran
ran; our contract draft was wrong, not the code
zhaoyuzhi/svcnet/GCS/generate_color_scribbles_ImageNet.py
code served (permissive licence) · get_code("2c0f9420e07c7b34")
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 list, ndim 1, 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 tempfile
import os

# Create a temporary text file with some content
with tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt') as f:
    temp_filename = f.name
    f.write("line one\n")
    f.write("line two\n")
    f.write("line three\n")

ARGS = {
    'filename': temp_filename
}
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 list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_confusion_matrix Not yet run
unverified (harvested, no recorded run)
zhaoyuzhi/svcnet/Evaluation/mIoU_Evaluation/svcnet_generate_anno.py
code served (permissive licence) · get_code("4b658dacb22598db")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 6 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 0 the same month, 0 later) and 5 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
text_readlines 2 other papers
2 dated earlier
1806.03589 (2018-06)
Free-Form Image Inpainting with Gated Convolution
2104.12357 (2021-04)
VCGAN: Video Colorization with Hybrid Generative Adversarial Network
PWCEstimate 1 other paper
1 dated earlier
2104.12357 (2021-04)
VCGAN: Video Colorization with Hybrid Generative Adversarial Network
Reshape_Tensor 1 other paper
1 dated earlier
2104.12357 (2021-04)
VCGAN: Video Colorization with Hybrid Generative Adversarial Network
get_confusion_matrix 1 other paper
1 dated earlier
1703.04730 (2017-03)
Understanding Black-box Predictions via Influence Functions
get_files 1 other paper
1 dated earlier
2104.12357 (2021-04)
VCGAN: Video Colorization with Hybrid Generative Adversarial Network
get_jpgs 1 other paper
1 dated earlier
2104.12357 (2021-04)
VCGAN: Video Colorization with Hybrid Generative Adversarial Network

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

In this paper, we propose a scribble-based video colorization network with temporal aggregation called SVCNet. It can colorize monochrome videos based on different user-given color scribbles. It addresses three common issues in the scribble-based video colorization area: colorization vividness, temporal consistency, and color bleeding. To improve the colorization quality and strengthen the temporal consistency, we adopt two sequential sub-networks in SVCNet for precise colorization and temporal smoothing, respectively. The first stage includes a pyramid feature encoder to incorporate color scribbles with a grayscale frame, and a semantic feature encoder to extract semantics. The second stage finetunes the output from the first stage by aggregating the information of neighboring colorized frames (as short-range connections) and the first colorized frame (as a long-range connection). To alleviate the color bleeding artifacts, we learn video colorization and segmentation simultaneously. Furthermore, we set the majority of operations on a fixed small image resolution and use a Super-resolution Module at the tail of SVCNet to recover original sizes. It allows the SVCNet to fit different image resolutions at the inference. Finally, we evaluate the proposed SVCNet on DAVIS and Videvo benchmarks. The experimental results demonstrate that SVCNet produces both higher-quality and more temporally consistent videos than other well-known video colorization approaches. The codes and models can be found at https://github.com/zhaoyuzhi/SVCNet.

For agents

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

get_harvested_code_for_paper("2303.11591")
get_code_for_paper("2303.11591")
have("2303.11591")

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

curl https://syntology.ai/api/ran/2303.11591.json

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

[![Syntology run record](https://syntology.ai/api/ran/2303.11591.svg)](https://syntology.ai/paper/2303.11591)

Connect an agent — have() is free.