SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.07486 · ECCV · 2024

OMR: Occlusion-Aware Memory-Based Renement for Video Lane Detection

Dongkwon Jin, Chang-Su Kim

arXiv · PDF · Open in the Atlas

Code that ran

5 of 6 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 1 has 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/2408.07486.json · badge: /api/ran/2408.07486.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
dongkwonjin/omr canonical (classifier label)
classifier basis: census
4 ran · 1 no recorded run
dongkwonjin/OMR community
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
Video_Memory Ran
ran on a synthesized input · constructed an object; no result was computed
dongkwonjin/OMR/Modeling/OpenLane-V/OMR/code/libs/video_memory.py
code served (permissive licence) · get_code("61f23a70edc69c49")
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, Video_Memory: 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 torch.nn as nn

torch.manual_seed(0)

class MockConfig:
    def __init__(self):
        self.num_t = 5

class MockModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.img_feat = torch.randn(2, 3, 4)
        self.obj_mask = torch.randn(2, 3, 4)
        self.curr_img_feat_lstm = torch.randn(2, 4)
        self.memory = None
        self.prev_frame_num = 0

cfg = MockConfig()

ARGS = {
    'cfg': 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 Video_Memory — 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 Video_Memory 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
call_culane_eval_custom Ran
ran; our contract draft was wrong, not the code
dongkwonjin/omr/Modeling/OpenLane-V/OMR/code/evaluation/evaluate_iou_official.py
code served (permissive licence) · get_code("5b1117c2c47c0666")
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 for the test
temp_dir = tempfile.mkdtemp()
data_dir = os.path.join(temp_dir, 'culane_data')
output_path = os.path.join(temp_dir, 'output')

# Create necessary directory structure
os.makedirs(os.path.join(data_dir, 'list', 'test_split'), exist_ok=True)
os.makedirs(output_path, exist_ok=True)

# Create test list files (empty files are sufficient for this setup)
list_files = [
    'test0_normal.txt',
    'test1_crowd.txt',
    'test2_hlight.txt',
    'test3_shadow.txt',
    'test4_noline.txt',
    'test5_arrow.txt',
    'test6_curve.txt',
    'test7_cross.txt',
    'test8_night.txt'
]

for list_file in list_files:
    list_path = os.path.join(data_dir, 'list', 'test_split', list_file)
    with open(list_path, 'w') as f:
        f.write('')

# Create mock output files with proper format for read_helper
output_txt_dir = os.path.join(output_path, 'txt')
os.makedirs(output_txt_dir, exist_ok=True)

output_files = [
    'out0_normal.txt',
    'out1_crowd.txt',
    'out2_hlight.txt',
    'out3_shadow.txt',
    'out4_noline.txt',
    'out5_arrow.txt',
    'out6_curve.txt',
    'out7_cross.txt',
    'out8_night.txt'
]

for out_file in output_files:
    out_path = os.path.join(output_txt_dir, out_file)
    with open(out_path, 'w') as f:
        # Write header and sample data that read_helper can parse
        f.write('Precision: 0.85 Recall: 0.90 F-measure: 0.875\n')

# Create a mock detect directory
detect_dir = os.path.join(output_path, 'exp_name')
os.makedirs(detect_dir, exist_ok=True)

ARGS = {
    'data_dir': data_dir,
    'exp_name': 'exp_name',
    'output_path': output_path,
    'category': 'test_all'
}
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
discrete_cross_iou Ran
ran; our contract draft was wrong, not the code
dongkwonjin/omr/Modeling/OpenLane-V/OMR/code/evaluation/evaluate_iou_laneatt.py
code served (permissive licence) · get_code("756bc2a2cd32caeb")
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 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 numpy as np

# Create small sample lanes for testing
# Each lane is a sequence of (x, y) points
lane1 = np.array([[100, 200], [150, 250], [200, 300]], dtype=np.float32)
lane2 = np.array([[120, 210], [170, 260], [220, 310]], dtype=np.float32)
lane3 = np.array([[500, 100], [550, 150], [600, 200]], dtype=np.float32)
lane4 = np.array([[520, 110], [570, 160], [620, 210]], dtype=np.float32)

xs = [lane1, lane2]
ys = [lane3, lane4]

ARGS = {
    "xs": xs,
    "ys": ys,
    "width": 30,
    "img_shape": (590, 1640, 3)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, 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
draw_lane Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
dongkwonjin/omr/Modeling/OpenLane-V/OMR/code/evaluation/evaluate_iou_laneatt.py
code served (permissive licence) · get_code("b274c90d2c686b65")
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 3, dtype family int, shape (100, 100, 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 cv2
import numpy as np

# Create a small lane array with coordinate pairs (x, y)
# Shape: (5, 2) - 5 points, each with x and y coordinate
lane = np.array([
    [10, 20],
    [15, 30],
    [20, 40],
    [25, 50],
    [30, 60]
], dtype=np.float32)

# Create an image with specific shape (100 height, 100 width, 3 channels for BGR)
img_shape = (100, 100, 3)
img = np.zeros(img_shape, dtype=np.uint8)

ARGS = {
    "lane": lane,
    "img": img,
    "img_shape": img_shape,
    "width": 30
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (100, 100, 3) — shape (100, 100, 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
read_helper Ran
ran; our contract draft was wrong, not the code
dongkwonjin/omr/Modeling/OpenLane-V/OMR/code/evaluation/evaluate_iou_official.py
code served (permissive licence) · get_code("a82bfd9d8b8b5cee")
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 tempfile
import os

# Create a temporary file with test data
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_path = temp_file.name

# Write test data in the format expected by read_helper
# The function expects: key1: value1 key2: value2 ...
# It skips the first line, then splits on spaces
# It assumes odd indices are values (1, 3, 5...) and even indices are keys (0, 2, 4...)
# Keys end with ':' which gets stripped

test_data = """header_line_to_skip
name: Alice age: 30 city: NYC score: 95.5
"""

temp_file.write(test_data)
temp_file.close()

ARGS = {
    'path': temp_path
}
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
call_culane_eval_custom_sampling Not yet run
unverified (harvested, no recorded run)
dongkwonjin/omr/Modeling/OpenLane-V/OMR/code/evaluation/evaluate_iou_official.py
code served (permissive licence) · get_code("e45a4247d7919703")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 0 later) and 3 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
call_culane_eval_custom 1 other paper
1 dated earlier
2203.15302 (2022-03)
Eigenlanes: Data-Driven Lane Descriptors for Structurally Diverse Lanes
draw_lane 1 other paper
1 dated earlier
2203.15302 (2022-03)
Eigenlanes: Data-Driven Lane Descriptors for Structurally Diverse Lanes
read_helper 1 other paper
1 dated earlier
2203.15302 (2022-03)
Eigenlanes: Data-Driven Lane Descriptors for Structurally Diverse Lanes

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

A novel algorithm for video lane detection is proposed in this paper. First, we extract a feature map for a current frame and detect a latent mask for obstacles occluding lanes. Then, we enhance the feature map by developing an occlusion-aware memory-based renement (OMR) module. It takes the obstacle mask and feature map from the current frame, previous output, and memory information as input, and processes them recursively in a video. Moreover, we apply a novel data augmentation scheme for training the OMR module eectively. Experimental results show that the proposed algorithm outperforms existing techniques on video lane datasets. Our codes are available at https://github.com/dongkwonjin/OMR.

For agents

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

get_harvested_code_for_paper("2408.07486")
get_code_for_paper("2408.07486")
have("2408.07486")

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

curl https://syntology.ai/api/ran/2408.07486.json

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

[![Syntology run record](https://syntology.ai/api/ran/2408.07486.svg)](https://syntology.ai/paper/2408.07486)

Connect an agent — have() is free.