SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · aaai_20232 · AAAI · 2022

Homography Decomposition Networks for Planar Object Tracking

Yang Li, Jianke Zhu, Xinrui Zhan, Yueran Liu

Open in the Atlas

Code that ran

9 of 14 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; 4 failures of our instrument; 5 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/aaai_20232.json · badge: /api/ran/aaai_20232.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
zhanxinrui/HDN canonical (classifier label)
classifier basis: census
9 ran · 5 no recorded run
FunctionStatusWhere it lives
check_keys Ran
ran on a synthesized input
zhanxinrui/HDN/hdn/utils/model_load.py
code served (permissive licence) · get_code("ec73a18eb6cee64f")
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 bool, 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
import torch.nn as nn
import logging

logger = logging.getLogger('global')

torch.manual_seed(0)

# Create a simple model with state_dict
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.layer1 = nn.Linear(4, 3)
        self.layer2 = nn.Linear(3, 2)
        self.bn = nn.BatchNorm1d(2)
    
    def forward(self, x):
        x = self.layer1(x)
        x = self.layer2(x)
        x = self.bn(x)
        return x

model = SimpleModel()

# Create a pretrained state dict with some overlapping and some unused keys
pretrained_state_dict = {
    'layer1.weight': torch.randn(3, 4),
    'layer1.bias': torch.randn(3),
    'layer2.weight': torch.randn(2, 3),
    'layer2.bias': torch.randn(2),
    'bn.weight': torch.randn(2),
    'bn.bias': torch.randn(2),
    'bn.running_mean': torch.randn(2),
    'bn.running_var': torch.randn(2),
    'bn.num_batches_tracked': torch.tensor(0),
    'unused_layer.weight': torch.randn(5, 5),
    'unused_layer.bias': torch.randn(5),
}

ARGS = {
    'model': model,
    'pretrained_state_dict': pretrained_state_dict,
}
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 bool, 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
conv3x3 Ran
ran; our contract draft was wrong, not the code
zhanxinrui/HDN/hdn/models/backbone/resnet_atrous.py
code served (permissive licence) · get_code("48468b2c75aae583")
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 Conv2d. 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,
    "stride": 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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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_1x1_bn Ran
ran; our contract draft was wrong, not the code
zhanxinrui/HDN/hdn/models/backbone/mobile_v2.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from zhoudaquan/rethinking_bottleneck_design, is pointer only (licence: NOASSERTION) · get_code("a0131fb70c267a9e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from zhoudaquan/rethinking_bottleneck_design, 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.

conv_bn Ran
ran on a synthesized input
zhanxinrui/HDN/hdn/models/backbone/mobile_v2.py
code served (permissive licence) · get_code("c46a07169e517d67")
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 Sequential. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'inp': 3,
    'oup': 16,
    'stride': 1,
    'padding': 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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_cls_loss Ran
ran on a synthesized input
zhanxinrui/HDN/hdn/models/loss.py
code served (permissive licence) · get_code("c15bf5953bf452e6")
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
import torch.nn.functional as F

torch.manual_seed(0)

# Create pred tensor: log probabilities for 2 samples, 4 classes
# F.nll_loss expects log probabilities as input
pred = torch.randn(2, 4)
pred = F.log_softmax(pred, dim=1)

# Create label tensor: class indices for 2 samples
label = torch.tensor([0, 2], dtype=torch.long)

# Create select tensor: indices to select from pred and label
select = torch.tensor([0, 1], dtype=torch.long)

ARGS = {
    "pred": pred,
    "label": label,
    "select": select
}
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
remove_prefix Ran
ran on a synthesized input
zhanxinrui/HDN/hdn/utils/model_load.py
code served (permissive licence) · get_code("a57313630ecbb3a1")
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 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 logging

logging.basicConfig(level=logging.INFO)

# Create a sample state_dict with some parameters that have a common prefix
state_dict = {
    'module.layer1.weight': torch.randn(4, 3),
    'module.layer1.bias': torch.randn(4),
    'module.layer2.weight': torch.randn(5, 4),
    'layer3.weight': torch.randn(3, 5),
}

# Define the prefix to remove
prefix = 'module.'

ARGS = {
    'state_dict': state_dict,
    'prefix': prefix,
}
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 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
select_cross_entropy_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zhanxinrui/HDN/hdn/models/loss.py
code served (permissive licence) · get_code("608039b454e6bd80")
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 float, ndim 0, dtype family float, shape () 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)

# Create pred tensor: shape (2, 2) for binary classification on 2 samples
# The function reshapes to (-1, 2), so we need a tensor with 2 as last dim
pred = torch.randn(2, 2, requires_grad=True)

# Create label tensor: shape (2,) with binary labels (0 or 1)
label = torch.tensor([1, 0], dtype=torch.long)

ARGS = {
    "pred": pred,
    "label": label,
}
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 float, 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
xcorr_depthwise Ran
ran; our contract draft was wrong, not the code
zhanxinrui/HDN/hdn/core/xcorr.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from hqucv/siamrn, is pointer only (licence: NONE) · get_code("541e18bc31d27b44")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from hqucv/siamrn, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

xcorr_fast Ran
ran on a synthesized input
zhanxinrui/HDN/hdn/core/xcorr.py
code served (permissive licence) · get_code("928a6daecaca5912")
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, 1, 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 torch.nn.functional as F

torch.manual_seed(0)

# x: input tensor with shape (batch, channels, height, width)
# The function calls:
# - kernel.size()[0] to get batch dimension
# - x.size()[1] to get channel dimension
# - kernel.size()[2], kernel.size()[3] to get kernel height/width
# - x.size()[2], x.size()[3] to get input height/width

# kernel: filter tensor with shape (batch, channels, kernel_h, kernel_w)
# The function reshapes it to (-1, x.size()[1], kernel.size()[2], kernel.size()[3])

# Create small tensors with distinct dimensions for clarity
batch_size = 2
in_channels = 4
kernel_h = 3
kernel_w = 3
input_h = 5
input_w = 5

x = torch.randn(batch_size, in_channels, input_h, input_w)
kernel = torch.randn(batch_size, in_channels, kernel_h, kernel_w)

ARGS = {
    "x": x,
    "kernel": kernel
}
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, 1, 3, 3) — shape (2, 1, 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
getLinearPolarImg Not yet run
unverified (harvested, no recorded run)
zhanxinrui/HDN/hdn/models/logpolar.py
code served (permissive licence) · get_code("b0e0e8e8d706ecdb")
How it ran

no invocation recorded.

getPolarImg Not yet run
unverified (harvested, no recorded run)
zhanxinrui/HDN/hdn/models/logpolar.py
code served (permissive licence) · get_code("ddab3e9ea434f29e")
How it ran

no invocation recorded.

load_pretrain Not yet run
unverified (harvested, no recorded run)
zhanxinrui/HDN/hdn/utils/model_load.py
code served (permissive licence) · get_code("e5561be5567ba02a")
How it ran

no invocation recorded.

select_xr_focal_fuse_smooth_l1_loss_top_k Not yet run
unverified (harvested, no recorded run)
zhanxinrui/HDN/hdn/models/loss.py
code served (permissive licence) · get_code("c49d1cc5867c98b8")
How it ran

no invocation recorded.

xcorr_slow Not yet run
unverified (harvested, no recorded run)
zhanxinrui/HDN/hdn/core/xcorr.py
code served (permissive licence) · get_code("74d343344c065353")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 10 are also present in other papers whose code we hold (10 with at least one paper dated earlier, 0 the same month, 4 later) and 4 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
conv_1x1_bn 31 other papers
19 dated earlier, 10 later, 2 same year, order not determinable · 2 only as an assembly brick
1703.07402 (2017-03)
Simple Online and Realtime Tracking with a Deep Association Metric
1801.04381 (2018-01)
MobileNetV2: Inverted Residuals and Linear Bottlenecks
1811.08383 (2018-11)
TSM: Temporal Shift Module for Efficient Video Understanding
and 28 more not named here
check_keys 4 other papers
2 dated earlier, 1 later, 1 same year, order not determinable
1812.11703 (2018-12)
SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks
2003.06761 (2020-03)
Siamese Box Adaptive Network for Visual Tracking
2203.10541 (2022-03)
Unsupervised Domain Adaptation for Nighttime Aerial Tracking
and 1 more not named here
conv3x3 4 other papers
2 dated earlier, 2 later
1812.11703 (2018-12)
SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks
2112.05006 (2021-12)
Exploring Event-driven Dynamic Context for Accident Scene Segmentation
2024.findings-emnlp.965 (2024 (year only))
See Detail Say Clear: Towards Brain CT Report Generation via Pathological Clue-driven Representation Learning
and 1 more not named here
get_cls_loss 4 other papers
3 dated earlier, 1 same year, order not determinable
1812.11703 (2018-12)
SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks
1911.07241 (2019-11)
SiamCAR: Siamese Fully Convolutional Classification and Regression for Visual Tracking
2003.06761 (2020-03)
Siamese Box Adaptive Network for Visual Tracking
and 1 more not named here
remove_prefix 4 other papers
2 dated earlier, 1 later, 1 same year, order not determinable
1812.11703 (2018-12)
SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks
2003.06761 (2020-03)
Siamese Box Adaptive Network for Visual Tracking
2203.10541 (2022-03)
Unsupervised Domain Adaptation for Nighttime Aerial Tracking
and 1 more not named here
select_cross_entropy_loss 4 other papers
3 dated earlier, 1 same year, order not determinable
1812.11703 (2018-12)
SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks
1911.07241 (2019-11)
SiamCAR: Siamese Fully Convolutional Classification and Regression for Visual Tracking
2003.06761 (2020-03)
Siamese Box Adaptive Network for Visual Tracking
and 1 more not named here
xcorr_depthwise 4 other papers
3 dated earlier, 1 same year, order not determinable
1812.11703 (2018-12)
SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks
2003.06761 (2020-03)
Siamese Box Adaptive Network for Visual Tracking
2104.00829 (2021-04)
Learning to Filter: Siamese Relation Network for Robust Tracking
and 1 more not named here
xcorr_fast 3 other papers
2 dated earlier, 1 same year, order not determinable
1812.11703 (2018-12)
SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks
2003.06761 (2020-03)
Siamese Box Adaptive Network for Visual Tracking
2203.10541 (2022-03)
Unsupervised Domain Adaptation for Nighttime Aerial Tracking
xcorr_slow 3 other papers
2 dated earlier, 1 same year, order not determinable
1812.11703 (2018-12)
SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks
2003.06761 (2020-03)
Siamese Box Adaptive Network for Visual Tracking
2203.10541 (2022-03)
Unsupervised Domain Adaptation for Nighttime Aerial Tracking
conv_bn 1 other paper
1 dated earlier
1812.11703 (2018-12)
SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep 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 2026-11-22.

Repositories linked to this paper

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

Abstract

Planar object tracking plays an important role in AI applications, such as robotics, visual servoing, and visual SLAM. Although the previous planar trackers work well in most scenarios, it is still a challenging task due to the rapid motion and large transformation between two consecutive frames. The essential reason behind this problem is that the condition number of such a non-linear system changes unstably when the searching range of the homography parameter space becomes larger. To this end, we propose a novel Homography Decomposition Networks (HDN) approach that drastically reduces and stabilizes the condition number by decomposing the homography transformation into two groups. Specifically, a similarity transformation estimator is designed to predict the first group robustly by a deep convolution equivariant network. By taking advantage of the scale and rotation estimation with high confidence, a residual transformation is estimated by a simple regression model. Furthermore, the proposed end-to-end network is trained in a semi-supervised fashion. Extensive experiments show that our proposed approach outperforms the state-of-the-art planar tracking methods at a large margin on the challenging POT, UCSB and POIC datasets. Codes and models are available at https://github.com/zhanxinrui/HDN.

For agents

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

get_harvested_code_for_paper("aaai_20232")
get_code_for_paper("aaai_20232")
have("aaai_20232")

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

curl https://syntology.ai/api/ran/aaai_20232.json

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

[![Syntology run record](https://syntology.ai/api/ran/aaai_20232.svg)](https://syntology.ai/paper/aaai_20232)

Connect an agent — have() is free.