SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2204.02574 · CVPR · 2022

FocalClick: Towards Practical Interactive Image Segmentation

Xi Chen, Hengshuang Zhao, Donglian Qi, Yilei Zhang, Manni Duan, Zhiyan Zhao

arXiv · PDF · Open in the Atlas

Code that ran

14 of 21 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 10 have no contract verdict; 4 failures of our instrument; 7 no recorded run), measured 2026-11-22; 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/2204.02574.json · badge: /api/ran/2204.02574.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
XavierCHEN34/ClickSEG marked official in Papers with Code
Papers with Code archive
14 ran · 6 no recorded run
copy not recorded community
not in the paper's declared repositories
0 ran · 1 no recorded run
FunctionStatusWhere it lives
BasicBlockV1b Ran
ran on a synthesized input
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("7e0099013f714d0b")
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 torch.nn as nn
import numpy as np

INIT_ARGS = {
    'inplanes': 4,
    'planes': 8,
    'stride': 1,
    'dilation': 2,
    'downsample': nn.Sequential(
        nn.Conv2d(4, 8, kernel_size=1, stride=1, bias=False),
        nn.BatchNorm2d(8)
    ),
    'previous_dilation': 1,
    'norm_layer': nn.BatchNorm2d
}

ARGS = {
    'x': torch.randn(2, 4, 6, 8)
}
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 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)
BatchImageNormalize Ran
ran on a synthesized input
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("339f5051599196e6")
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

INIT_ARGS = {
    'mean': [0.485, 0.456, 0.406, 0.5],
    'std': [0.229, 0.224, 0.225, 0.2],
    'dtype': torch.float,
}

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

class BatchImageNormalize:

    def __init__(self, mean, std, dtype=torch.float):
        self.mean = torch.as_tensor(mean, dtype=dtype)[None, :, None, None]
        self.std = torch.as_tensor(std, dtype=dtype)[None, :, None, None]

    def __call__(self, tensor):
        tensor = tensor.clone()
        tensor.sub_(self.mean.to(tensor.device)).div_(self.std.to(tensor.device))
        return tensor


if __name__ == '__main__':
    normalizer = BatchImageNormalize(**INIT_ARGS)
    result = normalizer(**ARGS)
    print(f"Input shape: {ARGS['tensor'].shape}")
    print(f"Output shape: {result.shape}")
    print(f"Output sample:\n{result}")
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)
BottleneckV1b Ran
ran on a synthesized input
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("d040bbf8b42c1853")
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, 64, 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
import torch.nn as nn

class BottleneckV1b(nn.Module):
    expansion = 4

    def __init__(self, inplanes, planes, stride=1, dilation=1, downsample=None, previous_dilation=1, norm_layer=nn.BatchNorm2d):
        super(BottleneckV1b, self).__init__()
        self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
        self.bn1 = norm_layer(planes)
        self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride, padding=dilation, dilation=dilation, bias=False)
        self.bn2 = norm_layer(planes)
        self.conv3 = nn.Conv2d(planes, planes * self.expansion, kernel_size=1, bias=False)
        self.bn3 = norm_layer(planes * self.expansion)
        self.relu = nn.ReLU(inplace=True)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x):
        residual = x
        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)
        out = self.conv2(out)
        out = self.bn2(out)
        out = self.relu(out)
        out = self.conv3(out)
        out = self.bn3(out)
        if self.downsample is not None:
            residual = self.downsample(x)
        out = out + residual
        out = self.relu(out)
        return out


# Create a simple downsample module for stride > 1
downsample_module = nn.Sequential(
    nn.Conv2d(16, 64, kernel_size=1, stride=2, bias=False),
    nn.BatchNorm2d(64)
)

INIT_ARGS = {
    'inplanes': 16,
    'planes': 16,
    'stride': 2,
    'dilation': 1,
    'downsample': downsample_module,
    'previous_dilation': 1,
    'norm_layer': nn.BatchNorm2d
}

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

if __name__ == '__main__':
    model = BottleneckV1b(**INIT_ARGS)
    model.eval()
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Expected shape: torch.Size([2, 64, 4, 4])")
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, 64, 4, 4) — shape (2, 64, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
ConvModule Ran
ran on a synthesized input
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("36ecf3e15546035e")
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, 16, 8, 8) 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

class ConvModule(nn.Module):

    def __init__(self, in_channels=32, out_channels=32, kernel_size=3, stride=1, padding=1):
        super().__init__()
        self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride, padding)
        self.norm = nn.BatchNorm2d(out_channels)
        self.activation = nn.ReLU()

    def forward(self, x):
        return self.activation(self.norm(self.conv(x)))


INIT_ARGS = {
    'in_channels': 3,
    'out_channels': 16,
    'kernel_size': 3,
    'stride': 1,
    'padding': 1,
}

ARGS = {
    'x': torch.randn(2, 3, 8, 8),
}

if __name__ == '__main__':
    module = ConvModule(**INIT_ARGS)
    output = module(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Expected shape: (2, 16, 8, 8)")
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, 16, 8, 8) — shape (2, 16, 8, 8) 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)
LRMult Ran
ran on a synthesized input
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("2028862f40603cc7")
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 NoneType. 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

class LRMult(object):

    def __init__(self, lr_mult=1.0):
        self.lr_mult = lr_mult

    def __call__(self, m):
        if getattr(m, 'weight', None) is not None:
            m.weight.lr_mult = self.lr_mult
        if getattr(m, 'bias', None) is not None:
            m.bias.lr_mult = self.lr_mult


INIT_ARGS = {
    'lr_mult': 0.1
}

# Create a simple module with weight and bias to pass to __call__
test_module = nn.Linear(4, 3)

ARGS = {
    'm': test_module
}

METHOD = '__call__'

# Construct and call
lr_mult_instance = LRMult(**INIT_ARGS)
lr_mult_instance(**ARGS)
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 NoneType — a NoneType output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a NoneType 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
ObjectAttentionBlock2D Ran
ran on a synthesized input
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("e0f1a4750642d91b")
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, 8, 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 as nn
from torch.nn import BatchNorm2d

INIT_ARGS = {
    'in_channels': 4,
    'key_channels': 8,
    'scale': 2,
    'norm_layer': BatchNorm2d,
    'align_corners': True,
}

ARGS = {
    'x': torch.randn(2, 4, 8, 6),
    'proxy': torch.randn(2, 4, 4, 3),
}
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, 8, 6) — shape (2, 4, 8, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
ScaleLayer Ran
ran on a synthesized input
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("7c22256846b5a788")
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 3, 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 numpy as np

# INIT_ARGS for ScaleLayer.__init__(init_value=1.0, lr_mult=1)
INIT_ARGS = {
    'init_value': 2.0,
    'lr_mult': 0.5
}

# ARGS for ScaleLayer.forward(x)
ARGS = {
    'x': torch.randn(2, 4, 3)
}
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 Tensor, ndim 3, 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)
SpatialGather_Module Ran
ran on a synthesized input
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("fc329aee6357b551")
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, 5, 1) 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
import torch.nn.functional as F

class SpatialGather_Module(nn.Module):
    """
        Aggregate the context features according to the initial
        predicted probability distribution.
        Employ the soft-weighted method to aggregate the context.
    """

    def __init__(self, cls_num=0, scale=1):
        super(SpatialGather_Module, self).__init__()
        self.cls_num = cls_num
        self.scale = scale

    def forward(self, feats, probs):
        batch_size, c, h, w = (probs.size(0), probs.size(1), probs.size(2), probs.size(3))
        probs = probs.view(batch_size, c, -1)
        feats = feats.view(batch_size, feats.size(1), -1)
        feats = feats.permute(0, 2, 1)
        probs = F.softmax(self.scale * probs, dim=2)
        ocr_context = torch.matmul(probs, feats).permute(0, 2, 1).unsqueeze(3)
        return ocr_context


INIT_ARGS = {
    'cls_num': 5,
    'scale': 2
}

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

if __name__ == '__main__':
    module = SpatialGather_Module(**INIT_ARGS)
    output = module(**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
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, 5, 1) — shape (2, 4, 5, 1) 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)
SpatialOCR_Module Ran
ran on a synthesized input
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("7936d6a591621025")
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 torch.nn as nn

# Import the classes
import torch.nn.functional as F

class ObjectAttentionBlock2D(nn.Module):
    """
    The basic implementation for object context block
    Input:
        N X C X H X W
    Parameters:
        in_channels       : the dimension of the input feature map
        key_channels      : the dimension after the key/query transform
        scale             : choose the scale to downsample the input feature maps (save memory cost)
        bn_type           : specify the bn type
    Return:
        N X C X H X W
    """

    def __init__(self, in_channels, key_channels, scale=1, norm_layer=nn.BatchNorm2d, align_corners=True):
        super(ObjectAttentionBlock2D, self).__init__()
        self.scale = scale
        self.in_channels = in_channels
        self.key_channels = key_channels
        self.align_corners = align_corners
        self.pool = nn.MaxPool2d(kernel_size=(scale, scale))
        self.f_pixel = nn.Sequential(nn.Conv2d(in_channels=self.in_channels, out_channels=self.key_channels, kernel_size=1, stride=1, padding=0, bias=False), nn.Sequential(norm_layer(self.key_channels), nn.ReLU(inplace=True)), nn.Conv2d(in_channels=self.key_channels, out_channels=self.key_channels, kernel_size=1, stride=1, padding=0, bias=False), nn.Sequential(norm_layer(self.key_channels), nn.ReLU(inplace=True)))
        self.f_object = nn.Sequential(nn.Conv2d(in_channels=self.in_channels, out_channels=self.key_channels, kernel_size=1, stride=1, padding=0, bias=False), nn.Sequential(norm_layer(self.key_channels), nn.ReLU(inplace=True)), nn.Conv2d(in_channels=self.key_channels, out_channels=self.key_channels, kernel_size=1, stride=1, padding=0, bias=False), nn.Sequential(norm_layer(self.key_channels), nn.ReLU(inplace=True)))
        self.f_down = nn.Sequential(nn.Conv2d(in_channels=self.in_channels, out_channels=self.key_channels, kernel_size=1, stride=1, padding=0, bias=False), nn.Sequential(norm_layer(self.key_channels), nn.ReLU(inplace=True)))
        self.f_up = nn.Sequential(nn.Conv2d(in_channels=self.key_channels, out_channels=self.in_channels, kernel_size=1, stride=1, padding=0, bias=False), nn.Sequential(norm_layer(self.in_channels), nn.ReLU(inplace=True)))

    def forward(self, x, proxy):
        batch_size, h, w = (x.size(0), x.size(2), x.size(3))
        if self.scale > 1:
            x = self.pool(x)
        query = self.f_pixel(x).view(batch_size, self.key_channels, -1)
        query = query.permute(0, 2, 1)
        key = self.f_object(proxy).view(batch_size, self.key_channels, -1)
        value = self.f_down(proxy).view(batch_size, self.key_channels, -1)
        value = value.permute(0, 2, 1)
        sim_map = torch.matmul(query, key)
        sim_map = self.key_channels ** (-0.5) * sim_map
        sim_map = F.softmax(sim_map, dim=-1)
        context = torch.matmul(sim_map, value)
        context = context.permute(0, 2, 1).contiguous()
        context = context.view(batch_size, self.key_channels, *x.size()[2:])
        context = self.f_up(context)
        if self.scale > 1:
            context = F.interpolate(input=context, size=(h, w), mode='bilinear', align_corners=self.align_corners)
        return context

class SpatialOCR_Module(nn.Module):
    """
    Implementation of the OCR module:
    We aggregate the global object representation to update the representation for each pixel.
    """

    def __init__(self, in_channels, key_channels, out_channels, scale=1, dropout=0.1, norm_layer=nn.BatchNorm2d, align_corners=True):
        super(SpatialOCR_Module, self).__init__()
        self.object_context_block = ObjectAttentionBlock2D(in_channels, key_channels, scale, norm_layer, align_corners)
        _in_channels = 2 * in_channels
        self.conv_bn_dropout = nn.Sequential(nn.Conv2d(_in_channels, out_channels, kernel_size=1, padding=0, bias=False), nn.Sequential(norm_layer(out_channels), nn.ReLU(inplace=True)), nn.Dropout2d(dropout))

    def forward(self, feats, proxy_feats):
        context = self.object_context_block(feats, proxy_feats)
        output = self.conv_bn_dropout(torch.cat([context, feats], 1))
        return output


INIT_ARGS = {
    'in_channels': 4,
    'key_channels': 2,
    'out_channels': 8,
    'scale': 1,
    'dropout': 0.1,
    'norm_layer': nn.BatchNorm2d,
    'align_corners': True,
}

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

if __name__ == '__main__':
    module = SpatialOCR_Module(**INIT_ARGS)
    output = module(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Expected shape: (2, 8, 4, 3)")
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)
XConvBnRelu Ran
ran on a synthesized input
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("44f751550a0af371")
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, 16, 4, 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 as nn

class XConvBnRelu(nn.Module):
    """
    Xception conv bn relu
    """

    def __init__(self, input_dims=3, out_dims=16, **kwargs):
        super(XConvBnRelu, self).__init__()
        self.conv3x3 = nn.Conv2d(input_dims, input_dims, 3, 1, 1, groups=input_dims)
        self.conv1x1 = nn.Conv2d(input_dims, out_dims, 1, 1, 0)
        self.norm = nn.BatchNorm2d(out_dims)
        self.activation = nn.ReLU()

    def forward(self, x):
        x = self.conv3x3(x)
        x = self.conv1x1(x)
        x = self.norm(x)
        x = self.activation(x)
        return x


INIT_ARGS = {
    'input_dims': 3,
    'out_dims': 16,
}

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

if __name__ == '__main__':
    model = XConvBnRelu(**INIT_ARGS)
    model.eval()
    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, 16, 4, 5) — shape (2, 16, 4, 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)
get_classname Ran
ran; our contract draft was wrong, not the code
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("532fac016c7c896a")
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 str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

class SimpleModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)

class CustomClass:
    pass

# Test with a built-in class
builtin_cls = int

# Test with a custom class
custom_cls = CustomClass

# Test with a module class
module_cls = SimpleModule

ARGS = {
    'cls': custom_cls
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_default_params Ran
ran; our contract draft was wrong, not the code
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("93ae2e1088f964ad")
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 dict. 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 inspect

torch.manual_seed(0)

# Create a simple test class hierarchy with default parameters
class BaseClass:
    def __init__(self, param1=10, param2="default"):
        self.param1 = param1
        self.param2 = param2

class DerivedClass(BaseClass, nn.Module):
    def __init__(self, param3=3.14, param1=5):
        nn.Module.__init__(self)
        BaseClass.__init__(self, param1=param1)
        self.param3 = param3

# The function takes a class as input
some_class = DerivedClass

ARGS = {
    "some_class": some_class
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
resize Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("55bc02c6846634c4")
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, 3, 8, 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
import torch.nn.functional as F

torch.manual_seed(0)

# Create a small input tensor for resizing
# Shape: (batch=2, channels=3, height=4, width=5)
input_tensor = torch.randn(2, 3, 4, 5)

# Define ARGS dict with all parameters
ARGS = {
    'input': input_tensor,
    'size': (8, 10),
    'scale_factor': None,
    'mode': 'nearest',
    'align_corners': None
}
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, 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
serialize Ran
ran; our contract draft was wrong, not the code
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("a4cd31544f3ea2b5")
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 function. 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 inspect
from functools import wraps
from copy import deepcopy

torch.manual_seed(0)

# Create a minimal nn.Module subclass to test with
class SimpleModule(nn.Module):
    def __init__(self, hidden_dim=64, dropout=0.1):
        super().__init__()
        self.linear = nn.Linear(hidden_dim, hidden_dim)
    
    def forward(self, x):
        return self.linear(x)

# Create a test function that uses serialize decorator
def test_init(self, input_size=10, output_size=5, learning_rate=0.001):
    self.input_size = input_size
    self.output_size = output_size
    self.learning_rate = learning_rate

# Get the serialize function (it's defined in the code above)
def get_classname(cls):
    module = cls.__module__
    name = cls.__qualname__
    if module is not None and module != '__builtin__':
        name = module + '.' + name
    return name

def get_default_params(some_class):
    params = dict()
    for mclass in some_class.mro():
        if mclass is nn.Module or mclass is object:
            continue
        mclass_params = inspect.signature(mclass.__init__).parameters
        for pname, param in mclass_params.items():
            if param.default != param.empty and pname not in params:
                params[pname] = param
    return params

def serialize(init):
    parameters = list(inspect.signature(init).parameters)

    @wraps(init)
    def new_init(self, *args, **kwargs):
        params = deepcopy(kwargs)
        for pname, value in zip(parameters[1:], args):
            params[pname] = value
        config = {'class': get_classname(self.__class__), 'params': dict()}
        specified_params = set(params.keys())
        for pname, param in get_default_params(self.__class__).items():
            if pname not in params:
                params[pname] = param.default
        for name, value in list(params.items()):
            param_type = 'builtin'
            if inspect.isclass(value):
                param_type = 'class'
                value = get_classname(value)
            config['params'][name] = {'type': param_type, 'value': value, 'specified': name in specified_params}
        setattr(self, '_config', config)
        init(self, *args, **kwargs)
    return new_init

# The serialize function takes one parameter: init (a function)
# We need to pass a valid __init__ method
ARGS = {
    'init': test_init
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
DistMaps Not yet run
unverified (harvested, no recorded run)
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("524aba093ebca5b5")
How it ran

no invocation recorded.

HRNetModel Not yet run
unverified (harvested, no recorded run)
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("5eac58acde5ed053")
How it ran

no invocation recorded.

HighResolutionModule Not yet run
unverified (harvested, no recorded run)
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("c3e49580d712bb6f")
How it ran

no invocation recorded.

HighResolutionNet Not yet run
unverified (harvested, no recorded run)
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("79f5c9927f655378")
How it ran

no invocation recorded.

ISModel Not yet run
unverified (harvested, no recorded run)
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("0467c4cb28e2a662")
How it ran

no invocation recorded.

RefineLayer Not yet run
unverified (harvested, no recorded run)
XavierCHEN34/ClickSEG/isegm/model/is_hrnet_model.py
code served (permissive licence) · get_code("f8c13074d5a56e32")
How it ran

no invocation recorded.

split_points_by_order Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from zmhhmz/gpcis_cvpr2023
pointer only · get_code("8bb23b71c17a0a25")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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 21 samples lifted from this paper's linked repositories, 21 carry a function identity and 0 do not. Of the 21 distinct functions with an identity, 5 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 3 later) and 16 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
split_points_by_order
brick of this paper's assembly: helper
4 other papers
4 later
2210.11006 (2022-10)
SimpleClick: Interactive Image Segmentation with Simple Vision Transformers
2302.14578 (2023-02)
Interactive Segmentation as Gaussian Process Classification
and 1 more not named here
resize
brick of this paper's assembly: helper
2 other papers
2 later
Du_Efficient_Mask_Correction_for_Click-Based_Interactive_Image_Segmentation_CVPR_2023_paper (2023 (year only))
Efficient Mask Correction for Click-Based Interactive Image Segmentation
2312.00752 (2023-12)
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
BasicBlockV1b
brick of this paper's assembly: helper
1 other paper
1 dated earlier
1802.02611 (2018-02)
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
BottleneckV1b
brick of this paper's assembly: helper
1 other paper
1 dated earlier
1802.02611 (2018-02)
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
SpatialGather_Module
brick of this paper's assembly: helper
1 other paper
1 later
2303.04249 (2023-03)
Where We Are and What We're Looking At: Query Based Worldwide Image Geo-localization Using Hierarchies and Scenes

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

Interactive segmentation allows users to extract target masks by making positive/negative clicks. Although explored by many previous works, there is still a gap between academic approaches and industrial needs: first, existing models are not efficient enough to work on lowpower devices; second, they perform poorly when used to refine preexisting masks as they could not avoid destroying the correct part. FocalClick solves both issues at once by predicting and updating the mask in localized areas. For higher efficiency, we decompose the slow prediction on the entire image into two fast inferences on small crops: a coarse segmentation on the Target Crop, and a local refinement on the Focus Crop. To make the model work with preexisting masks, we formulate a sub-task termed Interactive Mask Correction, and propose Progressive Merge as the solution. Progressive Merge exploits morphological information to decide where to preserve and where to update, enabling users to refine any preexisting mask effectively. FocalClick achieves competitive results against SOTA methods with significantly smaller FLOPs. It also shows significant superiority when making corrections on preexisting masks. Code and data will be released at github.com/XavierCHEN34/ClickSEG

For agents

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

get_harvested_code_for_paper("2204.02574")
get_code_for_paper("2204.02574")
have("2204.02574")

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

curl https://syntology.ai/api/ran/2204.02574.json

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

[![Syntology run record](https://syntology.ai/api/ran/2204.02574.svg)](https://syntology.ai/paper/2204.02574)

Connect an agent — have() is free.