SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2404.03015 · 2024

DPFT: Dual Perspective Fusion Transformer for Camera-Radar-based Object Detection

arXiv · PDF · Open in the Atlas

Code that ran

7 of 9 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 0 have no contract verdict; 7 failures of our instrument; 2 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/2404.03015.json · badge: /api/ran/2404.03015.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
tumftm/dpft canonical (classifier label)
classifier basis: census
7 ran · 2 no recorded run
FunctionStatusWhere it lives
build_scheduler Ran
ran; our contract draft was wrong, not the code
tumftm/dpft/src/dprt/training/scheduler.py
code served (permissive licence) · get_code("a9aeaa8ba5cade44")
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 partial. 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
from functools import partial

torch.manual_seed(0)

# Create a simple model and optimizer for the scheduler
model = nn.Linear(4, 3)
optimizer = torch.optim.SGD(model.parameters(), lr=0.1)

# Define arguments for build_scheduler
# Test with 'ChainedScheduler' name
name = 'ChainedScheduler'

# Prepare scheduler configurations as list of dicts
schedulers = [
    {'name': 'StepLR', 'step_size': 10, 'gamma': 0.1},
    {'name': 'ExponentialLR', 'gamma': 0.9}
]

# Build the ARGS dict with all parameters of build_scheduler
ARGS = {
    'name': name,
    'optimizer': optimizer,
    'schedulers': schedulers
}
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 partial — a partial output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a partial 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
chained_scheduler Ran
ran; our contract draft was wrong, not the code
tumftm/dpft/src/dprt/training/scheduler.py
code served (permissive licence) · get_code("f62ac1df15e6ba9a")
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 ChainedScheduler. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import torch.optim as optim

torch.manual_seed(0)

# Create a minimal model
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

model = SimpleModel()

# Create an optimizer
optimizer = optim.SGD(model.parameters(), lr=0.1)

# Create scheduler configurations
schedulers = [
    {
        'name': 'StepLR',
        'step_size': 10,
        'gamma': 0.1
    },
    {
        'name': 'ExponentialLR',
        'gamma': 0.9
    }
]

ARGS = {
    'optimizer': optimizer,
    'schedulers': schedulers
}
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 ChainedScheduler — a ChainedScheduler output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ChainedScheduler 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
focal_loss Ran
ran; our contract draft was wrong, not the code
tumftm/dpft/src/dprt/training/loss.py
code served (permissive licence) · get_code("abd213a9f852803b")
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 torch.nn.functional as F

torch.manual_seed(0)

# Create small tensors for focal_loss
batch_size = 2
feature_dim = 4

# inputs: logits from model predictions (B, feature_dim)
inputs = torch.randn(batch_size, feature_dim)

# targets: binary labels (B, feature_dim)
targets = torch.randint(0, 2, (batch_size, feature_dim)).float()

# alpha: weighting factor for positive vs negative examples
alpha = 0.75

# gamma: exponent for modulating factor
gamma = 2.0

# reduction: how to reduce the loss
reduction = 'none'

ARGS = {
    'inputs': inputs,
    'targets': targets,
    'alpha': alpha,
    'gamma': gamma,
    'reduction': reduction
}
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 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
get_box_corners Ran
ran; our contract draft was wrong, not the code
tumftm/dpft/src/dprt/utils/bbox.py
code served (permissive licence) · get_code("9eccf35752a7240f")
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, 4, 8, 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

torch.manual_seed(0)

# Create small tensors with distinct dimensions
# B=2 (batch), N=4 (number of boxes), spatial dims as needed
B = 2
N = 4

center = torch.randn(B, N, 3)
size = torch.randn(B, N, 3).abs() + 0.5  # ensure positive sizes
angle = torch.randn(B, N)

ARGS = {
    "center": center,
    "size": size,
    "angle": angle,
}
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, 4, 8, 3) — shape (2, 4, 8, 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
get_box_volume_from_corners Ran
ran; our contract draft was wrong, not the code
tumftm/dpft/src/dprt/utils/bbox.py
code served (permissive licence) · get_code("858633a79b1fb482")
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 1, dtype family float, shape (2,) 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)

# Create a batch of 2 bounding boxes, each with 8 corners in 3D space
# Shape: (N, 8, 3) where N=2, 8 corners, 3 coordinates (x, y, z)
boxes = torch.tensor([
    [
        [0.0, 0.0, 0.0],  # corner 0
        [1.0, 0.0, 0.0],  # corner 1
        [1.0, 0.0, 1.0],  # corner 2
        [0.0, 1.0, 0.0],  # corner 3
        [0.0, 0.0, 2.0],  # corner 4
        [1.0, 0.0, 2.0],  # corner 5
        [1.0, 1.0, 2.0],  # corner 6
        [0.0, 1.0, 2.0],  # corner 7
    ],
    [
        [0.0, 0.0, 0.0],  # corner 0
        [2.0, 0.0, 0.0],  # corner 1
        [2.0, 0.0, 1.0],  # corner 2
        [0.0, 3.0, 0.0],  # corner 3
        [0.0, 0.0, 4.0],  # corner 4
        [2.0, 0.0, 4.0],  # corner 5
        [2.0, 3.0, 4.0],  # corner 6
        [0.0, 3.0, 4.0],  # corner 7
    ]
], dtype=torch.float32)

ARGS = {
    "boxes": boxes,
}
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 1, dtype family float, shape (2,) — shape (2,) 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_config Ran
ran; our contract draft was wrong, not the code
tumftm/dpft/src/dprt/utils/config.py
code served (permissive licence) · get_code("065ee82d0907c020")
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 json
import tempfile
import os

# Create a temporary JSON configuration file
temp_dir = tempfile.mkdtemp()
config_file = os.path.join(temp_dir, "config.json")

# Write a sample configuration to the file
sample_config = {
    "model_name": "test_model",
    "learning_rate": 0.001,
    "batch_size": 32,
    "epochs": 10,
    "layers": [64, 128, 256]
}

with open(config_file, 'w') as f:
    json.dump(sample_config, f)

ARGS = {
    "file": config_file
}
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
loads_config Ran
ran; our contract draft was wrong, not the code
tumftm/dpft/src/dprt/utils/config.py
code served (permissive licence) · get_code("eefe0c3768f91d17")
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 json

def loads_config(file: str) -> dict:
    """Load a configuration file.

    Arguments:
        file: A serialized json file.

    Returns:
        Configuration dictionary.
    """
    config = json.loads(file)
    return config


# Create a valid JSON string as input
config_dict = {
    "model_name": "test_model",
    "learning_rate": 0.001,
    "batch_size": 32,
    "num_epochs": 10,
    "layers": [64, 128, 256]
}
json_file_content = json.dumps(config_dict)

ARGS = {
    "file": json_file_content
}
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
get_minimum_enclosing_box_corners Not yet run
unverified (harvested, no recorded run)
tumftm/dpft/src/dprt/utils/bbox.py
code served (permissive licence) · get_code("37dbd12763e70cc3")
How it ran

no invocation recorded.

sequential_lr Not yet run
unverified (harvested, no recorded run)
tumftm/dpft/src/dprt/training/scheduler.py
code served (permissive licence) · get_code("5f5714501b6da323")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 9 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.

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 perception of autonomous vehicles has to be efficient, robust, and cost-effective. However, cameras are not robust against severe weather conditions, lidar sensors are expensive, and the performance of radar-based perception is still inferior to the others. Camera-radar fusion methods have been proposed to address this issue, but these are constrained by the typical sparsity of radar point clouds and often designed for radars without elevation information. We propose a novel camera-radar fusion approach called Dual Perspective Fusion Transformer (DPFT), designed to overcome these limitations. Our method leverages lower-level radar data (the radar cube) instead of the processed point clouds to preserve as much information as possible and employs projections in both the camera and ground planes to effectively use radars with elevation information and simplify the fusion with camera data. As a result, DPFT has demonstrated state-of-the-art performance on the K-Radar dataset while showing remarkable robustness against adverse weather conditions and maintaining a low inference time. The code is made available as open-source software under https://github.com/TUMFTM/DPFT.

For agents

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

get_harvested_code_for_paper("2404.03015")
get_code_for_paper("2404.03015")
have("2404.03015")

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

curl https://syntology.ai/api/ran/2404.03015.json

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

[![Syntology run record](https://syntology.ai/api/ran/2404.03015.svg)](https://syntology.ai/paper/2404.03015)

Connect an agent — have() is free.