SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2104.14545 · CVPR · 2021

LightTrack: Finding Lightweight Neural Networks for Object Tracking via One-Shot Architecture Search

Houwen Peng, Jianlong Fu, Dong Wang, Huchuan Lu, Kan Wu, Bin Yan

arXiv · PDF · Open in the Atlas

Code that ran

7 of 7 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 3 failures of our instrument; 0 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/2104.14545.json · badge: /api/ran/2104.14545.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
researchmm/LightTrack marked official in Papers with Code
Papers with Code archive
7 ran · 0 no recorded run
FunctionStatusWhere it lives
Config Ran
ran on a synthesized input · constructed an object; no result was computed
researchmm/LightTrack/lib/tracker/lighttrack.py
code served (permissive licence) · get_code("a36dcbde2714611a")
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, Config: 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 numpy as np

torch.manual_seed(0)

# The Config class takes two optional parameters: stride and even
# Based on the __init__ signature: def __init__(self, stride=8, even=0)

ARGS = {
    'stride': 8,
    'even': 0
}
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 Config — 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 Config 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
Lighttrack Ran
ran on a synthesized input · constructed an object; no result was computed
researchmm/LightTrack/lib/tracker/lighttrack.py
code served (permissive licence) · get_code("bf42cff83b1cbcf3")
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, Lighttrack: the run constructed the object and did not call it, so no result was computed. 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 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

torch.manual_seed(0)

class DummyModel(nn.Module):
    def __init__(self):
        super(DummyModel, self).__init__()
        self.conv = nn.Conv2d(3, 16, 3)
    
    def forward(self, x):
        return self.conv(x)

class DummyInfo:
    def __init__(self):
        self.stride = 8
        self.dataset = 'OTB'

# Create minimal valid arguments
info = DummyInfo()
even = 0

ARGS = {
    'info': info,
    'even': even
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy, torch, yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Lighttrack — 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 Lighttrack 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_subwindow_tracking Ran
ran; our contract draft was wrong, not the code
researchmm/LightTrack/lib/tracker/lighttrack.py
code served (permissive licence) · get_code("a5063c71c5931d27")
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 numpy as np
import torch
import cv2

torch.manual_seed(0)

# Create a small test image (height=100, width=120, channels=3)
im = np.random.randint(0, 256, size=(100, 120, 3), dtype=np.uint8)

# Position - center of the image
pos = [60.0, 50.0]

# Model size and original size for cropping
model_sz = 127
original_sz = 255

# Average channel values (for padding)
avg_chans = np.array([104, 117, 123], dtype=np.uint8)

# Output mode
out_mode = 'torch'

ARGS = {
    'im': im,
    'pos': pos,
    'model_sz': model_sz,
    'original_sz': original_sz,
    'avg_chans': avg_chans,
    'out_mode': out_mode
}
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 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
im_to_torch Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
researchmm/LightTrack/lib/tracker/lighttrack.py
code served (permissive licence) · get_code("25c9237afd263361")
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 3, dtype family float, shape (2, 3, 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 numpy as np
import torch

torch.manual_seed(0)

# Create a small test image: (height=3, width=4, channels=2)
# Using shape (H, W, C) format as input
img = np.random.rand(3, 4, 2).astype(np.float32)

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 3, dtype family float, shape (2, 3, 4) — shape (2, 3, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_yaml Ran
ran; our contract draft was wrong, not the code
researchmm/LightTrack/lib/tracker/lighttrack.py
code served (permissive licence) · get_code("4b0995306a51fa38")
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 yaml
import tempfile
import os

# Create a temporary YAML file with the expected structure
yaml_content = """
TEST:
  learning_rate: 0.001
  batch_size: 32
  epochs: 10
  model_name: "test_model"
OTHER:
  some_param: "value"
"""

# Write to a temporary file
temp_file = tempfile.NamedTemporaryFile(mode='w', suffix='.yaml', delete=False)
temp_file.write(yaml_content)
temp_file.close()

ARGS = {
    'path': temp_file.name,
    'subset': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — 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
python2round Ran
ran, honoured the contract we drafted
researchmm/LightTrack/lib/tracker/lighttrack.py
code served (permissive licence) · get_code("80095723929304cb")
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. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function python2round takes a single parameter 'f' which should be a float
# It uses basic arithmetic operations on f (addition, subtraction, multiplication, division)
# and the built-in round() function

ARGS = {
    "f": 2.5
}
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
to_torch Ran
ran, honoured the contract we drafted
researchmm/LightTrack/lib/tracker/lighttrack.py
code served (permissive licence) · get_code("516e33dfd5c4f7df")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. 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)

# to_torch takes an ndarray and converts it to a torch tensor
# It only calls torch.from_numpy on the input
ndarray = np.array([[1.0, 2.0, 3.0, 4.0],
                     [5.0, 6.0, 7.0, 8.0]], dtype=np.float32)

ARGS = {
    "ndarray": ndarray
}
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 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 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
im_to_torch
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2007.09115 (2020-07)
Scale Equivariance Improves Siamese Tracking
to_torch
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2007.09115 (2020-07)
Scale Equivariance Improves Siamese Tracking

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

Object tracking has achieved significant progress over the past few years. However, state-of-the-art trackers become increasingly heavy and expensive, which limits their deployments in resource-constrained applications. In this work, we present LightTrack, which uses neural architecture search (NAS) to design more lightweight and efficient object trackers. Comprehensive experiments show that our LightTrack is effective. It can find trackers that achieve superior performance compared to handcrafted SOTA trackers, such as SiamRPN++ [30] and Ocean [56], while using much fewer model Flops and parameters. Moreover, when deployed on resource-constrained mobile chipsets, the discovered trackers run much faster. For example, on Snapdragon 845 Adreno GPU, LightTrack runs 12× faster than Ocean, while using 13× fewer parameters and 38× fewer Flops. Such improvements might narrow the gap between academic models and industrial deployments in object tracking task. LightTrack is released at here.

For agents

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

get_harvested_code_for_paper("2104.14545")
get_code_for_paper("2104.14545")
have("2104.14545")

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

curl https://syntology.ai/api/ran/2104.14545.json

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

[![Syntology run record](https://syntology.ai/api/ran/2104.14545.svg)](https://syntology.ai/paper/2104.14545)

Connect an agent — have() is free.