SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.00969 · 2024

Visible-Thermal Multiple Object Tracking: Large-scale Video Dataset and Progressive Fusion Approach

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 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; 1 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 4 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2408.00969.json · badge: /api/ran/2408.00969.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
wqw123wqw/pftrack canonical (classifier label)
classifier basis: census
6 ran · 1 no recorded run
FunctionStatusWhere it lives
geometric_mean Ran
ran, violated the contract we drafted
wqw123wqw/pftrack/trackeval/plotting.py
code served (permissive licence) · get_code("06d2be4ac9d0e95b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 1, dtype family float, shape (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

# The function geometric_mean takes two parameters x and y
# and computes their geometric mean using numpy operations
# Both parameters should be numpy arrays or numeric values

x = np.array([1.0, 4.0, 9.0, 16.0])
y = np.array([4.0, 9.0, 16.0, 25.0])

ARGS = {
    'x': x,
    'y': y
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (4,) — shape (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
init_config Ran
ran; our contract draft was wrong, not the code
wqw123wqw/pftrack/trackeval/utils.py
code served (permissive licence) · get_code("0ce058e665602da2")
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

torch.manual_seed(0)

# Analyze the function:
# - config: dict, accessed with .keys() and [] indexing, may be None
# - default_config: dict, accessed with .keys() and [] indexing
# - name: optional string, used in conditional and print
# The function checks if config is None, merges keys from default_config,
# and optionally prints based on config['PRINT_CONFIG']

config = {
    'learning_rate': 0.001,
    'batch_size': 32,
    'epochs': 10
}

default_config = {
    'learning_rate': 0.0001,
    'batch_size': 16,
    'epochs': 5,
    'PRINT_CONFIG': False,
    'optimizer': 'adam'
}

name = 'TestModel'

ARGS = {
    'config': config,
    'default_config': default_config,
    'name': name
}
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
jaccard Ran
ran, honoured the contract we drafted
wqw123wqw/pftrack/trackeval/plotting.py
code served (permissive licence) · get_code("00c662d6d9709fe4")
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)

# The function takes two parameters x and y
# Both are used in arithmetic operations (division, multiplication, addition, subtraction)
# They should be numeric types (tensors, arrays, or scalars)

# Create small distinct tensors for x and y
x = torch.tensor([[1.0, 2.0, 3.0, 4.0],
                   [5.0, 6.0, 7.0, 8.0]])

y = torch.tensor([[10.0, 20.0, 30.0, 40.0],
                   [50.0, 60.0, 70.0, 80.0]])

ARGS = {
    "x": x,
    "y": y
}
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
load_multiple_tracker_summaries Ran
ran; our contract draft was wrong, not the code
wqw123wqw/pftrack/trackeval/plotting.py
code served (permissive licence) · get_code("f33c76e72d8a6635")
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 os
import tempfile
import shutil

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

# Create tracker folders and summary files
tracker_names = ['tracker_1', 'tracker_2']
for tracker in tracker_names:
    tracker_path = os.path.join(temp_dir, tracker)
    os.makedirs(tracker_path, exist_ok=True)
    
    # Create a summary file for class 'pedestrian'
    summary_file = os.path.join(tracker_path, 'pedestrian_summary.txt')
    with open(summary_file, 'w') as f:
        f.write('MOTA MOTP Precision Recall\n')
        f.write('0.75 0.82 0.88 0.92\n')

ARGS = {
    'tracker_folder': temp_dir,
    'tracker_list': ['tracker_1', 'tracker_2'],
    'cls': 'pedestrian'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
time Ran
ran; our contract draft was wrong, not the code
wqw123wqw/pftrack/trackeval/_timing.py
code served (permissive licence) · get_code("df1bd31f4da1bed6")
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 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 inspect
from functools import wraps
from time import perf_counter

# Set up the timing decorator and global state
DO_TIMING = False
DISPLAY_LESS_PROGRESS = False
timer_dict = {}
counter = 0

def time(f):
    @wraps(f)
    def wrap(*args, **kw):
        if DO_TIMING:
            ts = perf_counter()
            result = f(*args, **kw)
            te = perf_counter()
            tt = te - ts
            arg_names = inspect.getfullargspec(f)[0]
            if arg_names[0] == 'self' and DISPLAY_LESS_PROGRESS:
                return result
            elif arg_names[0] == 'self':
                method_name = type(args[0]).__name__ + '.' + f.__name__
            else:
                method_name = f.__name__
            if method_name in timer_dict.keys():
                timer_dict[method_name] += tt
            else:
                timer_dict[method_name] = tt
            if method_name == 'Evaluator.evaluate':
                print('')
                print('Timing analysis:')
                for key, value in timer_dict.items():
                    print('%-70s %2.4f sec' % (key, value))
            else:
                arg_titles = ['tracker', 'seq', 'cls']
                arg_vals = []
                for i, a in enumerate(arg_names):
                    if a in arg_titles:
                        arg_vals.append(args[i])
                arg_text = '(' + ', '.join(arg_vals) + ')'
                if arg_names[0] == 'self':
                    print('%-74s %2.4f sec' % (' ' * 4 + method_name + arg_text, tt))
                elif arg_names[0] == 'test':
                    pass
                else:
                    global counter
                    counter += 1
                    print('%i %-70s %2.4f sec' % (counter, method_name + arg_text, tt))
            return result
        else:
            return f(*args, **kw)
    return wrap

# Test function that uses the decorator
@time
def sample_function(x, y):
    return x + y

@time
def sample_method_function(self, tracker, seq, cls):
    return tracker + seq + cls

class DummyClass:
    @time
    def sample_method(self, tracker, seq, cls):
        return tracker + seq + cls

# The `time` function itself takes one parameter: `f` (a function to be decorated)
# So we need to provide a callable function as the argument

def dummy_func(a, b):
    """A simple function to be decorated."""
    return a + b

ARGS = {
    'f': dummy_func
}
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 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
update_config Ran
ran; our contract draft was wrong, not the code
wqw123wqw/pftrack/trackeval/utils.py
code served (permissive licence) · get_code("5ef3f617efb9ddc4")
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 argparse
import sys

# Create a sample config dict with various types of values
config = {
    'learning_rate': 0.001,
    'batch_size': 32,
    'epochs': 100,
    'use_cuda': False,
    'model_name': 'resnet50',
    'layers': [64, 128, 256],
    'checkpoint': None,
}

# Simulate command line arguments
sys.argv = [
    'script.py',
    '--learning_rate', '0.01',
    '--batch_size', '64',
    '--use_cuda', 'False',
    '--model_name', 'vgg16',
    '--layers', '128', '256', '512',
]

ARGS = {
    'config': config,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
validate_metrics_list Not yet run
unverified (harvested, no recorded run)
wqw123wqw/pftrack/trackeval/utils.py
code served (permissive licence) · get_code("9792bf1150d36891")
How it ran

no invocation recorded.

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, 7 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 0 the same month, 4 later) and 0 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
geometric_mean 6 other papers
5 dated earlier, 1 later
1905.04804 (2019-05)
Video Instance Segmentation
1912.07515 (2019-12)
Learning a Neural Solver for Multiple Object Tracking
2009.07736 (2020-09)
HOTA: A Higher Order Metric for Evaluating Multi-Object Tracking
and 3 more not named here
jaccard 6 other papers
5 dated earlier, 1 later
1905.04804 (2019-05)
Video Instance Segmentation
1912.07515 (2019-12)
Learning a Neural Solver for Multiple Object Tracking
2009.07736 (2020-09)
HOTA: A Higher Order Metric for Evaluating Multi-Object Tracking
and 3 more not named here
load_multiple_tracker_summaries 6 other papers
5 dated earlier, 1 later
1905.04804 (2019-05)
Video Instance Segmentation
1912.07515 (2019-12)
Learning a Neural Solver for Multiple Object Tracking
2009.07736 (2020-09)
HOTA: A Higher Order Metric for Evaluating Multi-Object Tracking
and 3 more not named here
time 6 other papers
5 dated earlier, 1 later
1905.04804 (2019-05)
Video Instance Segmentation
1912.07515 (2019-12)
Learning a Neural Solver for Multiple Object Tracking
2009.07736 (2020-09)
HOTA: A Higher Order Metric for Evaluating Multi-Object Tracking
and 3 more not named here
init_config 4 other papers
4 dated earlier
1905.04804 (2019-05)
Video Instance Segmentation
1912.07515 (2019-12)
Learning a Neural Solver for Multiple Object Tracking
2009.07736 (2020-09)
HOTA: A Higher Order Metric for Evaluating Multi-Object Tracking
and 1 more not named here
update_config 4 other papers
4 dated earlier
1905.04804 (2019-05)
Video Instance Segmentation
1912.07515 (2019-12)
Learning a Neural Solver for Multiple Object Tracking
2009.07736 (2020-09)
HOTA: A Higher Order Metric for Evaluating Multi-Object Tracking
and 1 more not named here
validate_metrics_list 4 other papers
4 dated earlier
1905.04804 (2019-05)
Video Instance Segmentation
1912.07515 (2019-12)
Learning a Neural Solver for Multiple Object Tracking
2009.07736 (2020-09)
HOTA: A Higher Order Metric for Evaluating Multi-Object Tracking
and 1 more not named here

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

The complementary benefits from visible and thermal infrared data are widely utilized in various computer vision task, such as visual tracking, semantic segmentation and object detection, but rarely explored in Multiple Object Tracking (MOT). In this work, we contribute a large-scale Visible-Thermal video benchmark for MOT, called VT-MOT. VT-MOT has the following main advantages. 1) The data is large scale and high diversity. VT-MOT includes 582 video sequence pairs, 401k frame pairs from surveillance, drone, and handheld platforms. 2) The cross-modal alignment is highly accurate. We invite several professionals to perform both spatial and temporal alignment frame by frame. 3) The annotation is dense and high-quality. VT-MOT has 3.99 million annotation boxes annotated and double-checked by professionals, including heavy occlusion and object re-acquisition (object disappear and reappear) challenges. To provide a strong baseline, we design a simple yet effective tracking framework, which effectively fuses temporal information and complementary information of two modalities in a progressive manner, for robust visible-thermal MOT. A comprehensive experiment are conducted on VT-MOT and the results prove the superiority and effectiveness of the proposed method compared with state-of-the-art methods. From the evaluation results and analysis, we specify several potential future directions for visible-thermal MOT. The project is released in https://github.com/wqw123wqw/PFTrack.

For agents

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

get_harvested_code_for_paper("2408.00969")
get_code_for_paper("2408.00969")
have("2408.00969")

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

curl https://syntology.ai/api/ran/2408.00969.json

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

[![Syntology run record](https://syntology.ai/api/ran/2408.00969.svg)](https://syntology.ai/paper/2408.00969)

Connect an agent — have() is free.