SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2308.04583 · ICCV · 2023

LATR: 3D Lane Detection from Monocular Images with Transformer

Shuguang Cui, Zhen Li, Chaoda Zheng, Sse, Xu Yan, Fnii, Yueru Luo, Tang Kun, Tencent Map, T Lab

arXiv · PDF · Open in the Atlas

Code that ran

11 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 10 failures of our instrument; 3 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/2308.04583.json · badge: /api/ran/2308.04583.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
JMoonr/LATR canonical (classifier label)
classifier basis: census
11 ran · 3 no recorded run
FunctionStatusWhere it lives
aligned_bilinear Ran
ran; our contract draft was wrong, not the code
JMoonr/LATR/models/sparse_inst_loss.py
code served (permissive licence) · get_code("29dbbaafa4be18e1")
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, 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 4D tensor with shape (batch, channels, height, width)
# Using distinct sizes: batch=2, channels=3, height=4, width=5
tensor = torch.randn(2, 3, 4, 5)

# Factor must be >= 1 and an integer
factor = 2

ARGS = {
    "tensor": tensor,
    "factor": factor,
}
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, 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
build_ms2one Ran
ran; our contract draft was wrong, not the code
JMoonr/LATR/models/ms2one.py
code served (permissive licence) · get_code("deccd45b122f35df")
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 Naive. 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 copy

torch.manual_seed(0)

# Test case 1: Naive configuration
config_naive = {
    'type': 'Naive',
    'inc': 4,
    'outc': 8,
    'kernel_size': 1
}

# Test case 2: DilateNaive configuration with single input channel
config_dilate_naive_single = {
    'type': 'DilateNaive',
    'inc': 4,
    'outc': 8,
    'num_scales': 4,
    'dilations': (1, 2, 5, 9),
    'merge': True,
    'fpn': False,
    'target_shape': None,
    'one_layer_before': False
}

# Test case 3: DilateNaive configuration with list of input channels
config_dilate_naive_multi = {
    'type': 'DilateNaive',
    'inc': [4, 4, 4, 4],
    'outc': 8,
    'num_scales': 4,
    'dilations': (1, 2, 5, 9),
    'merge': True,
    'fpn': True,
    'target_shape': (16, 16),
    'one_layer_before': True
}

ARGS = {
    'config': config_naive
}
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 Naive — a Naive output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Naive 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
build_nn_loss Ran
ran; our contract draft was wrong, not the code
JMoonr/LATR/models/latr_head.py
code served (permissive licence) · get_code("4c0969530211fe04")
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 CrossEntropyLoss. 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)

loss_cfg = {
    'type': 'CrossEntropyLoss',
    'reduction': 'mean'
}

ARGS = {
    'loss_cfg': loss_cfg
}
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 CrossEntropyLoss — a CrossEntropyLoss output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a CrossEntropyLoss 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
inverse_sigmoid Ran
ran; our contract draft was wrong, not the code
JMoonr/LATR/models/utils.py
code served (permissive licence) · get_code("c12fa9dce0d5fd34")
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

torch.manual_seed(0)

# Create a small tensor with values between 0 and 1 (valid sigmoid output range)
x = torch.tensor([[0.2, 0.5, 0.8, 0.1],
                   [0.9, 0.3, 0.7, 0.4]], dtype=torch.float32)

# Define arguments dictionary
ARGS = {
    'x': x,
    'eps': 1e-05
}
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
nested_masks_from_list Ran
ran; our contract draft was wrong, not the code
JMoonr/LATR/models/sparse_inst_loss.py
code served (permissive licence) · get_code("70f2a7785acb5193")
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 NestedTensor. 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
from typing import Optional, List

torch.manual_seed(0)

# Create a list of 3D tensors with different spatial sizes
# Shape: (channels, height, width)
tensor1 = torch.randn(4, 5, 6)  # 4 channels, 5x6 spatial
tensor2 = torch.randn(4, 3, 4)  # 4 channels, 3x4 spatial
tensor3 = torch.randn(4, 7, 8)  # 4 channels, 7x8 spatial

tensor_list = [tensor1, tensor2, tensor3]

ARGS = {
    "tensor_list": tensor_list,
    "input_shape": None
}
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 NestedTensor — a NestedTensor output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a NestedTensor 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
nested_tensor_from_tensor_list Ran
ran; our contract draft was wrong, not the code
JMoonr/LATR/models/sparse_inst_loss.py
code served (permissive licence) · get_code("e0e4b133a453e3c3")
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 NestedTensor. 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
from typing import List

torch.manual_seed(0)

# Create a list of 3D tensors with different shapes
# Each tensor has shape (C, H, W) where C=channels, H=height, W=width
tensor1 = torch.randn(3, 4, 5)  # 3 channels, 4 height, 5 width
tensor2 = torch.randn(3, 6, 7)  # 3 channels, 6 height, 7 width

tensor_list = [tensor1, tensor2]

ARGS = {
    "tensor_list": tensor_list
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type NestedTensor — a NestedTensor output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a NestedTensor 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
pos2posemb3d Ran
ran; our contract draft was wrong, not the code
JMoonr/LATR/models/transformer_bricks.py
code served (permissive licence) · get_code("615e86e6f80e97ba")
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, 384) 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 math

torch.manual_seed(0)

# Create a small 3D position tensor
# Shape: (batch_size, 3) where 3 represents x, y, z coordinates
pos = torch.randn(2, 3, dtype=torch.float32)

# Define arguments
ARGS = {
    "pos": pos,
    "num_pos_feats": 128,
    "temperature": 10000
}
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, 384) — shape (2, 384) 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
reduce_tensor Ran
ran; our contract draft was wrong, not the code
JMoonr/LATR/experiments/ddp.py
code served (permissive licence) · get_code("be5abd3083ea4db8")
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.distributed as dist
from unittest.mock import MagicMock, patch

torch.manual_seed(0)

# Mock dist.all_reduce to avoid actual distributed setup
original_all_reduce = dist.all_reduce

def mock_all_reduce(tensor, op=None):
    # Simulate reduction by summing across mock processes
    # For testing purposes, just return without error
    pass

# Patch the distributed module
dist.all_reduce = mock_all_reduce
dist.ReduceOp.SUM = MagicMock()

# Create a small tensor (batch=2, dim=4)
tensor = torch.randn(2, 4)

# Set world_size to a reasonable distributed training value
world_size = 4

ARGS = {
    "tensor": tensor,
    "world_size": world_size,
}
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
scatter_add Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
JMoonr/LATR/models/scatter_utils.py
code served (permissive licence) · get_code("e2d07b9f31178c8b")
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, 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 small test tensors
src = torch.tensor([[1.0, 2.0], [3.0, 4.0]])  # shape (2, 2)
index = torch.tensor([[0, 1], [0, 1]], dtype=torch.long)  # shape (2, 2)
dim = -1
out = None
dim_size = None

ARGS = {
    "src": src,
    "index": index,
    "dim": dim,
    "out": out,
    "dim_size": dim_size,
}
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, 2) — shape (2, 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
scatter_sum Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
JMoonr/LATR/models/scatter_utils.py
code served (permissive licence) · get_code("db8f88398b43049d")
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, 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)

# Construct arguments for scatter_sum function
src = torch.randn(2, 4)  # source tensor with shape (2, 4)
index = torch.tensor([[0, 1, 0, 2], [1, 0, 2, 1]], dtype=torch.long)  # index tensor matching src shape
dim = 1  # dimension to scatter along
out = None  # optional output tensor, None means create new
dim_size = 3  # optional size of dimension after scatter

ARGS = {
    "src": src,
    "index": index,
    "dim": dim,
    "out": out,
    "dim_size": dim_size,
}
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, 3) — shape (2, 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
to_python_float Ran
ran, honoured the contract we drafted
JMoonr/LATR/experiments/ddp.py
pointer only (licence: MIT); the stored copy get_code serves, from lengsicong/tell2design, is pointer only (licence: NONE) · get_code("6b5d363ab4dbf012")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from lengsicong/tell2design, 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.

broadcast Not yet run
unverified (harvested, no recorded run)
JMoonr/LATR/models/scatter_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from EdwardBerman/rayleigh_analysis, is pointer only (licence: NONE) · get_code("5dd38da29766376b")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from EdwardBerman/rayleigh_analysis, 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. No invocation recorded.

generate_ref_pt Not yet run
unverified (harvested, no recorded run)
JMoonr/LATR/models/transformer_bricks.py
code served (permissive licence) · get_code("190869ab9960b5ad")
How it ran

no invocation recorded.

ground2img Not yet run
unverified (harvested, no recorded run)
JMoonr/LATR/models/transformer_bricks.py
code served (permissive licence) · get_code("2d041f7c6096d5d9")
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, 2 the same month, 9 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
nested_tensor_from_tensor_list 55 other papers
29 dated earlier, 1 the same month, 24 later, 1 same year, order not determinable · 17 only as an assembly brick
1608.03983 (2016-08)
SGDR: Stochastic Gradient Descent with Warm Restarts
2005.12872 (2020-05)
End-to-End Object Detection with Transformers
2101.01909 (2021-01)
Line Segment Detection Using Transformers without Edges
and 52 more not named here
inverse_sigmoid 51 other papers
28 dated earlier, 3 the same month, 20 later · 23 only as an assembly brick
2005.12872 (2020-05)
End-to-End Object Detection with Transformers
2010.04159 (2020-10)
Deformable DETR: Deformable Transformers for End-to-End Object Detection
2103.11161 (2021-03)
MonteFloor: Extending MCTS for Reconstructing Accurate Large-Scale Floor Plans
and 48 more not named here
aligned_bilinear 17 other papers
9 dated earlier, 7 later, 1 same year, order not determinable · 1 only as an assembly brick
2201.00487 (2022-01)
Language as Queries for Referring Video Object Segmentation
2203.12827 (2022-03)
Sparse Instance Activation for Real-Time Instance Segmentation
2206.03789 (2022-06)
Language-Bridged Spatial-Temporal Interaction for Referring Video Object Segmentation
and 14 more not named here
to_python_float 9 other papers
7 dated earlier, 2 later
1904.00962 (2019-04)
Large Batch Optimization for Deep Learning: Training BERT in 76 minutes
1909.03329 (2019-09)
LAMOL: LAnguage MOdeling for Lifelong Language Learning
2109.04332 (2021-09)
PPT: Pre-trained Prompt Tuning for Few-shot Learning
and 6 more not named here
pos2posemb3d 6 other papers
3 dated earlier, 3 later
2206.01256 (2022-06)
PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images
2207.13080 (2022-07)
DETRs with Hybrid Matching
2304.01054 (2023-04)
VoxelFormer: Bird's-Eye-View Feature Generation based on Dual-view Attention for Multi-view 3D Object Detection
and 3 more not named here
broadcast 4 other papers
1 dated earlier, 3 later
2204.10555 (2022-04)
KALA: Knowledge-Augmented Language Model Adaptation
2311.16716 (2023-11)
GraphPro: Graph Pre-training and Prompt Learning for Recommendation
2506.16110 (2025-06)
Mitigating Over-Squashing in Graph Neural Networks by Spectrum-Preserving Sparsification
and 1 more not named here
scatter_sum 3 other papers
1 dated earlier, 2 later
2204.10555 (2022-04)
KALA: Knowledge-Augmented Language Model Adaptation
2311.16716 (2023-11)
GraphPro: Graph Pre-training and Prompt Learning for Recommendation
2602.05352 (2026-02)
Smoothness Errors in Dynamics Models and How to Avoid Them
reduce_tensor 2 other papers
1 dated earlier, 1 later
2106.06819 (2021-06)
D2C: Diffusion-Denoising Models for Few-shot Conditional Generation
2310.15171 (2023-10)
RoboDepth: Robust Out-of-Distribution Depth Estimation under Corruptions
scatter_add 2 other papers
1 dated earlier, 1 later
2204.10555 (2022-04)
KALA: Knowledge-Augmented Language Model Adaptation
2311.16716 (2023-11)
GraphPro: Graph Pre-training and Prompt Learning for Recommendation
nested_masks_from_list 1 other paper
1 dated earlier
2203.12827 (2022-03)
Sparse Instance Activation for Real-Time Instance Segmentation

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

3D lane detection from monocular images is a fundamental yet challenging task in autonomous driving. Recent advances primarily rely on structural 3D surrogates (e.g., bird's eye view) built from front-view image features and camera parameters. However, the depth ambiguity in monocular images inevitably causes misalignment between the constructed surrogate feature map and the original image, posing a great challenge for accurate lane detection. To address the above issue, we present a novel LATR model, an end-to-end 3D lane detector that uses 3Daware front-view features without transformed view representation. Specifically, LATR detects 3D lanes via crossattention based on query and key-value pairs, constructed using our lane-aware query generator and dynamic 3D ground positional embedding. On the one hand, each query is generated based on 2D lane-aware features and adopts a hybrid embedding to enhance the lane information. On the other hand, 3D space information is injected as positional embedding from an iteratively-updated 3D ground plane. LATR outperforms previous state-of-the-art methods on both synthetic Apollo and realistic OpenLane and ONCE-3DLanes by large margins (e.g., 11.4 gain in terms of F1 score on OpenLane). Code will be released at https://github.com/JMoonr/LATR.

For agents

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

get_harvested_code_for_paper("2308.04583")
get_code_for_paper("2308.04583")
have("2308.04583")

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

curl https://syntology.ai/api/ran/2308.04583.json

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

[![Syntology run record](https://syntology.ai/api/ran/2308.04583.svg)](https://syntology.ai/paper/2308.04583)

Connect an agent — have() is free.