SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2108.03151 · ICCV · 2021

Full-Duplex Strategy for Video Object Segmentation

Ling Shao, Zhe Wu, Jianbing Shen, Deng-Ping Fan, Keren Fu, Ge-Peng Ji

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (6 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 7 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-21; 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/2108.03151.json · badge: /api/ran/2108.03151.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
GewelsJI/FSNet marked official in Papers with Code
Papers with Code archive
7 ran · 1 no recorded run
FunctionStatusWhere it lives
BPM Ran
ran on a synthesized input
GewelsJI/FSNet/lib/fsnet.py
code served (permissive licence) · get_code("396efed75928c79c")
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 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 as nn
import numpy as np

INIT_ARGS = {
    'channel': 4
}

# Create input tensors with distinct shapes for different scales
# x_s1, x_s2, x_s3, x_s4 (spatial scales)
# x_e1, x_e2, x_e3, x_e4 (edge scales)
ARGS = {
    'x_s1': torch.randn(2, 4, 8, 6),  # batch=2, channel=4, H=8, W=6
    'x_s2': torch.randn(2, 4, 6, 4),  # batch=2, channel=4, H=6, W=4
    'x_s3': torch.randn(2, 4, 4, 3),  # batch=2, channel=4, H=4, W=3
    'x_s4': torch.randn(2, 4, 3, 2),  # batch=2, channel=4, H=3, W=2
    'x_e1': torch.randn(2, 4, 8, 6),  # same spatial size as x_s1
    'x_e2': torch.randn(2, 4, 6, 4),  # same spatial size as x_s2
    'x_e3': torch.randn(2, 4, 4, 3),  # same spatial size as x_s3
    'x_e4': torch.randn(2, 4, 3, 2)   # same spatial size as x_s4
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
BasicConv2d Ran
ran on a synthesized input · constructed an object; no result was computed
GewelsJI/FSNet/lib/fsnet.py
code served (permissive licence) · get_code("9d65fd2b9de9e302")
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 an instance of its own class, BasicConv2d: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'in_planes': 3,
    'out_planes': 64,
    'kernel_size': 3,
    'stride': 1,
    'padding': 1,
    'dilation': 1
}
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 BasicConv2d — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a BasicConv2d 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
Decoder Ran
ran on a synthesized input · constructed an object; no result was computed
GewelsJI/FSNet/lib/fsnet.py
code served (permissive licence) · get_code("7d142ec657be7a17")
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 an instance of its own class, Decoder: the run constructed the object and did not call it, so no result was computed. 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)

class BasicConv2d(nn.Module):
    def __init__(self, in_planes, out_planes, kernel_size, stride=1, padding=0, dilation=1):
        super(BasicConv2d, self).__init__()
        self.conv_bn = nn.Sequential(
            nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, padding=padding, dilation=dilation, bias=False),
            nn.BatchNorm2d(out_planes)
        )

    def forward(self, x):
        x = self.conv_bn(x)
        return x

class PyramidPooling(nn.Module):
    def __init__(self, in_channels, out_channels, **kwargs):
        super(PyramidPooling, self).__init__()
        inter_channels = int(in_channels / 4)
        self.conv1 = BasicConv2d(in_channels, inter_channels, 1, **kwargs)
        self.conv2 = BasicConv2d(in_channels, inter_channels, 1, **kwargs)
        self.conv3 = BasicConv2d(in_channels, inter_channels, 1, **kwargs)
        self.conv4 = BasicConv2d(in_channels, inter_channels, 1, **kwargs)
        self.out = BasicConv2d(in_channels * 2, out_channels, 1)

    def pool(self, x, size):
        avgpool = nn.AdaptiveAvgPool2d(size)
        return avgpool(x)

    def upsample(self, x, size):
        import torch.nn.functional as F
        return F.interpolate(x, size, mode='bilinear', align_corners=True)

    def forward(self, x):
        size = x.size()[2:]
        feat1 = self.upsample(self.conv1(self.pool(x, 1)), size)
        feat2 = self.upsample(self.conv2(self.pool(x, 2)), size)
        feat3 = self.upsample(self.conv3(self.pool(x, 3)), size)
        feat4 = self.upsample(self.conv4(self.pool(x, 6)), size)
        x = torch.cat([x, feat1, feat2, feat3, feat4], dim=1)
        x = self.out(x)
        return x

class Decoder(nn.Module):
    def __init__(self, channel):
        super(Decoder, self).__init__()
        self.upsample = nn.Upsample(scale_factor=2, mode='bilinear', align_corners=True)
        self.conv_upsample1 = BasicConv2d(channel, channel, 3, padding=1)
        self.conv_upsample2 = BasicConv2d(channel, channel, 3, padding=1)
        self.conv_upsample3 = BasicConv2d(channel, channel, 3, padding=1)
        self.ppm1 = PyramidPooling(channel, channel)
        self.ppm2 = PyramidPooling(channel, channel)
        self.ppm3 = PyramidPooling(channel, channel)
        self.ppm4 = PyramidPooling(channel, channel)
        self.conv_cat1 = nn.Sequential(
            BasicConv2d(2 * channel, 2 * channel, 3, padding=1),
            BasicConv2d(2 * channel, channel, 1)
        )
        self.conv_cat2 = nn.Sequential(
            BasicConv2d(2 * channel, 2 * channel, 3, padding=1),
            BasicConv2d(2 * channel, channel, 1)
        )
        self.conv_cat3 = nn.Sequential(
            BasicConv2d(2 * channel, 2 * channel, 3, padding=1),
            BasicConv2d(2 * channel, channel, 1)
        )
        self.output = nn.Sequential(
            BasicConv2d(channel, channel, 3, padding=1),
            nn.Conv2d(channel, 1, 1)
        )

    def forward(self, x1, x2, x3, x4):
        x4 = self.ppm4(x4)
        x3 = torch.cat((x3, self.conv_upsample1(self.upsample(x4))), 1)
        x3 = self.conv_cat1(x3)
        x3 = self.ppm3(x3)
        x2 = torch.cat((x2, self.conv_upsample2(self.upsample(x3))), 1)
        x2 = self.conv_cat2(x2)
        x2 = self.ppm2(x2)
        x1 = torch.cat((x1, self.conv_upsample3(self.upsample(x2))), 1)
        x1 = self.conv_cat3(x1)
        x1 = self.ppm1(x1)
        x = self.output(x1)
        return x

# Create arguments for Decoder
channel = 4

ARGS = {
    'channel': channel
}
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 Decoder — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Decoder 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
DimReduce Ran
ran on a synthesized input · constructed an object; no result was computed
GewelsJI/FSNet/lib/fsnet.py
code served (permissive licence) · get_code("b87d230de5341d9c")
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 an instance of its own class, DimReduce: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "in_channel": 4,
    "out_channel": 8,
}
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 DimReduce — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a DimReduce 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
PyramidPooling Ran
ran on a synthesized input · constructed an object; no result was computed
GewelsJI/FSNet/lib/fsnet.py
code served (permissive licence) · get_code("dfbf76d4ebcee7c3")
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 an instance of its own class, PyramidPooling: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'in_channels': 64,
    'out_channels': 128,
}
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 PyramidPooling — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a PyramidPooling 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
ResNet50 Ran
ran on a synthesized input · constructed an object; no result was computed
GewelsJI/FSNet/lib/fsnet.py
code served (permissive licence) · get_code("a42e339cce788ad4")
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 an instance of its own class, ResNet50: the run constructed the object and did not call it, so no result was computed. 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
import math

torch.manual_seed(0)

# ResNet50 takes no arguments in __init__
ARGS = {}
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 ResNet50 — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ResNet50 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
conv_upsample Ran
ran on a synthesized input · constructed an object; no result was computed
GewelsJI/FSNet/lib/fsnet.py
code served (permissive licence) · get_code("5d8bdcbcbac6daf1")
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 an instance of its own class, conv_upsample: the run constructed the object and did not call it, so no result was computed. 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
import torch.nn.functional as F

torch.manual_seed(0)

class BasicConv2d(nn.Module):
    def __init__(self, in_planes, out_planes, kernel_size, stride=1, padding=0, dilation=1):
        super(BasicConv2d, self).__init__()
        self.conv_bn = nn.Sequential(nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, padding=padding, dilation=dilation, bias=False), nn.BatchNorm2d(out_planes))

    def forward(self, x):
        x = self.conv_bn(x)
        return x

class conv_upsample(nn.Module):
    def __init__(self, channel):
        super(conv_upsample, self).__init__()
        self.conv = BasicConv2d(channel, channel, 1)

    def forward(self, x, target):
        if x.size()[2:] != target.size()[2:]:
            x = self.conv(F.upsample(x, size=target.size()[2:], mode='bilinear', align_corners=True))
        return x

# Create a conv_upsample instance with channel=4
channel = 4
model = conv_upsample(channel)

# Create test tensors
# x: batch_size=2, channels=4, height=8, width=6
x = torch.randn(2, 4, 8, 6)
# target: batch_size=2, channels=4, height=16, width=12 (different spatial dimensions to trigger upsample)
target = torch.randn(2, 4, 16, 12)

ARGS = {
    "channel": channel
}
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 conv_upsample — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a conv_upsample 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
FSNet Not yet run
unverified (harvested, no recorded run)
GewelsJI/FSNet/lib/fsnet.py
code served (permissive licence) · get_code("710857c7b9a46774")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 later) and 7 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
BasicConv2d
brick of this paper's assembly: helper
1 other paper
1 dated earlier · 1 only as an assembly brick
2101.10241 (2021-01)
RGB-D Salient Object Detection via 3D Convolutional Neural Networks

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Appearance and motion are two important sources of information in video object segmentation (VOS). Previous methods mainly focus on using simplex solutions, lowering the upper bound of feature collaboration among and across these two cues. In this paper, we study a novel framework, termed the FSNet (Full-duplex Strategy Network), which designs a relational cross-attention module (RCAM) to achieve the bidirectional message propagation across embedding subspaces. Furthermore, the bidirectional purification module (BPM) is introduced to update the inconsistent features between the spatial-temporal embeddings, effectively improving the model robustness. By considering the mutual restraint within the full-duplex strategy, our FS-Net performs the cross-modal feature-passing (i.e., transmission and receiving) simultaneously before the fusion and decoding stage, making it robust to various challenging scenarios (e.g., motion blur, occlusion) in VOS. Extensive experiments on five popular benchmarks (i.e., DAVIS 16 , FBMS, MCL, SegTrack-V2, and DAVSOD 19 ) show that our FSNet outperforms other state-of-the-arts for both the VOS and video salient object detection tasks.

For agents

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

get_harvested_code_for_paper("2108.03151")
get_code_for_paper("2108.03151")
have("2108.03151")

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

curl https://syntology.ai/api/ran/2108.03151.json

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

[![Syntology run record](https://syntology.ai/api/ran/2108.03151.svg)](https://syntology.ai/paper/2108.03151)

Connect an agent — have() is free.