SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2002.08394 · 2020

MonoLayout: Amodal scene layout from a single image

arXiv · PDF · Open in the Atlas

Code that ran

9 of 12 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 7 have no contract verdict; 2 failures of our instrument; 3 no recorded run), measured 2026-11-21; 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/2002.08394.json · badge: /api/ran/2002.08394.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
hbutsuak95/monolayout community (classifier label)
classifier basis not recorded on the link
9 ran · 3 no recorded run
FunctionStatusWhere it lives
get_lane_bev Ran
ran on a synthesized input
hbutsuak95/monolayout/preprocessing/argoverse/generate_groundtruth.py
code served (permissive licence) · get_code("2ebdc695607c310e")
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 (256, 256, 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 numpy as np
import cv2

class MockSE3:
    """Mock SE3 transformation class"""
    def inverse(self):
        return self
    
    def transform_point_cloud(self, points):
        # Simple identity transformation - just return points as-is
        return points

class MockAVM:
    """Mock Argoverse map object"""
    def get_lane_segment_polygon(self, lane_id, city_name):
        # Return a simple polygon as nx3 array (x, y, z coordinates)
        # Create a small rectangular polygon
        polygon = np.array([
            [0.0, 0.0, 5.0],
            [2.0, 0.0, 5.0],
            [2.0, 0.0, 10.0],
            [0.0, 0.0, 10.0]
        ], dtype=np.float32)
        return polygon

# Create mock objects
avm = MockAVM()
city_name = "test_city"
city_to_egovehicle_se3 = MockSE3()
egovehicle_SE3_cam = MockSE3()

# Create a small top-view image (256x256 with 3 channels for BGR)
top_view = np.zeros((256, 256, 3), dtype=np.uint8)

# Arguments from get_args()
base_path = './kitti_raw'
out_dir = ''
range_val = 40
occ_map_size = 256
seg_class = 'road'

# Calculate resolution
res = range_val / float(occ_map_size)

# Lane color in BGR format
lane_colour = (0, 255, 0)  # Green

# Lane IDs to process
req_lane_ids = [1, 2]

ARGS = {
    'req_lane_ids': req_lane_ids,
    'avm': avm,
    'city_name': city_name,
    'city_to_egovehicle_se3': city_to_egovehicle_se3,
    'egovehicle_SE3_cam': egovehicle_SE3_cam,
    'top_view': top_view,
    'res': res,
    'lane_colour': lane_colour
}
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 (256, 256, 3) — shape (256, 256, 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_lane_bev Ran
ran on a synthesized input
hbutsuak95/monolayout/preprocessing/argoverse/generate_weak_supervision.py
code served (permissive licence) · get_code("d5b31fbd0ece44e5")
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 (256, 256, 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 numpy as np
import cv2

class MockSE3:
    """Mock SE3 transformation class"""
    def __init__(self):
        pass
    
    def inverse(self):
        return self
    
    def transform_point_cloud(self, points):
        # Return points with some transformation applied
        # points should be shape (N, 3)
        return points + np.random.randn(*points.shape) * 0.01

class MockAVM:
    """Mock Argoverse Map object"""
    def __init__(self):
        pass
    
    def get_lane_segment_polygon(self, lane_id, city_name):
        # Return a polygon as (N, 3) array of 3D points
        num_points = 8
        polygon = np.random.randn(num_points, 3) * 10
        return polygon.astype(np.float32)

# Set up mock objects
avm = MockAVM()
city_name = "PaloAlto"
city_to_egovehicle_se3 = MockSE3()
egovehicle_SE3_cam = MockSE3()

# Create a small top-view image (BEV)
top_view = np.zeros((256, 256, 3), dtype=np.uint8)

# Resolution calculation
res = 40.0 / 256.0

# Lane colour in BGR
lane_colour = (0, 255, 0)

# Request lane IDs
req_lane_ids = [0, 1]

ARGS = {
    'req_lane_ids': req_lane_ids,
    'avm': avm,
    'city_name': city_name,
    'city_to_egovehicle_se3': city_to_egovehicle_se3,
    'egovehicle_SE3_cam': egovehicle_SE3_cam,
    'top_view': top_view,
    'res': res,
    'lane_colour': lane_colour
}
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 (256, 256, 3) — shape (256, 256, 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_pixel_area Ran
ran on a synthesized input
hbutsuak95/monolayout/utils.py
code served (permissive licence) · get_code("bdafeec9ac4bebf3")
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 int, ndim 0, dtype family int, 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 accesses segm.shape[0] and segm.shape[1]
# So segm needs to be an object with a shape attribute that has at least 2 dimensions
# A torch tensor or numpy array works perfectly

segm = torch.randn(4, 3)

ARGS = {
    "segm": segm
}
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 int, ndim 0, dtype family int, 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
get_rect Ran
ran on a synthesized input
hbutsuak95/monolayout/preprocessing/kitti/generate_supervision.py
code served (permissive licence) · get_code("f911f918dca05031")
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 (5, 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

ARGS = {
    'x': 10.0,
    'y': 15.0,
    'width': 4.0,
    'height': 6.0,
    'theta': np.pi / 4
}
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 2, dtype family float, shape (5, 2) — shape (5, 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
mean_IU Ran
ran on a synthesized input
hbutsuak95/monolayout/utils.py
code served (permissive licence) · get_code("bf70ece0cb49b07f")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 simple segmentation masks for testing
# eval_segm: predicted segmentation
# gt_segm: ground truth segmentation
# Both should be 2D arrays with class labels

height, width = 4, 3
np.random.seed(0)

# Create ground truth segmentation with classes 0, 1, 2
gt_segm = np.array([
    [0, 0, 1],
    [0, 1, 1],
    [2, 2, 1],
    [2, 2, 2]
], dtype=np.int32)

# Create predicted segmentation with some differences
eval_segm = np.array([
    [0, 0, 1],
    [0, 0, 1],
    [2, 1, 1],
    [2, 2, 2]
], dtype=np.int32)

ARGS = {
    "eval_segm": eval_segm,
    "gt_segm": gt_segm
}
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 list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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
mean_precision Ran
ran on a synthesized input
hbutsuak95/monolayout/utils.py
code served (permissive licence) · get_code("35beb490387990bc")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 math

# Create simple segmentation maps
# eval_segm: predicted segmentation
# gt_segm: ground truth segmentation
# Both should be 2D arrays with class labels

height = 4
width = 3

# Ground truth segmentation - contains classes 0, 1, 2
gt_segm = np.array([
    [0, 0, 1],
    [0, 1, 1],
    [2, 2, 1],
    [2, 2, 1]
], dtype=np.int32)

# Evaluated segmentation - same size, similar but not identical predictions
eval_segm = np.array([
    [0, 0, 1],
    [0, 0, 1],
    [2, 1, 1],
    [2, 2, 2]
], dtype=np.int32)

ARGS = {
    'eval_segm': eval_segm,
    'gt_segm': gt_segm
}
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 list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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
pil_loader Ran
ran on a synthesized input
hbutsuak95/monolayout/monolayout/datasets.py
code served (permissive licence) · get_code("e20df18137875803")
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 Image. 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 PIL.Image as pil
import io
import torch

# Create a minimal valid image file in memory
def create_test_image():
    """Create a small RGB image and return its bytes."""
    img = pil.new('RGB', (4, 3), color=(255, 0, 0))
    buffer = io.BytesIO()
    img.save(buffer, format='PNG')
    return buffer.getvalue()

# Write the image bytes to a temporary file path
import tempfile
import os

temp_dir = tempfile.mkdtemp()
test_image_path = os.path.join(temp_dir, 'test_image.png')

with open(test_image_path, 'wb') as f:
    f.write(create_test_image())

ARGS = {
    'path': test_image_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
resnet_multiimage_input Ran
ran; our contract draft was wrong, not the code
hbutsuak95/monolayout/monolayout/resnet_encoder.py
code served (permissive licence) · get_code("1ddc4a13b2c2af3d")
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 ResNetMultiImageInput. 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 torchvision.models as models

torch.manual_seed(0)

ARGS = {
    "num_layers": 18,
    "pretrained": False,
    "num_input_images": 1,
}
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 ResNetMultiImageInput — a ResNetMultiImageInput output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNetMultiImageInput 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
upsample Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
hbutsuak95/monolayout/monolayout/model.py
pointer only (licence: MIT); the stored copy get_code serves, from leofansq/CBR, is pointer only (licence: NONE) · get_code("a739072b350093e7")
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 leofansq/CBR, 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.

get_road_velo_pts Not yet run
unverified (harvested, no recorded run)
hbutsuak95/monolayout/preprocessing/kitti/generate_supervision.py
code served (permissive licence) · get_code("ee57e65c9fc60d65")
How it ran

no invocation recorded.

process_topview Not yet run
unverified (harvested, no recorded run)
hbutsuak95/monolayout/monolayout/datasets.py
code served (permissive licence) · get_code("d8a3cf14e8d15652")
How it ran

no invocation recorded.

resize_topview Not yet run
unverified (harvested, no recorded run)
hbutsuak95/monolayout/monolayout/datasets.py
code served (permissive licence) · get_code("5682838d74b48e82")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 2 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.

FunctionAlso present inEarliest-datedOthers named
upsample 8 other papers
1 dated earlier, 7 later
1912.03426 (2019-12)
Self-Supervised 3D Keypoint Learning for Ego-motion Estimation
2103.14201 (2021-03)
Image2Reverb: Cross-Modal Reverb Impulse Response Synthesis
2108.08829 (2021-08)
Fine-grained Semantics-aware Representation Enhancement for Self-supervised Monocular Depth Estimation
and 5 more not named here
resnet_multiimage_input 6 other papers
6 later
Zhou_R-MSFM_Recurrent_Multi-Scale_Feature_Modulation_for_Monocular_Depth_Estimating_ICCV_2021_paper (2021 (year only))
R-MSFM: Recurrent Multi-Scale Feature Modulation for Monocular Depth Estimating
2301.01006 (2023-01)
Policy Pre-training for Autonomous Driving via Self-supervised Geometric Modeling
2309.05254 (2023-09)
Towards Better Data Exploitation in Self-Supervised Monocular Depth Estimation
and 3 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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

In this paper, we address the novel, highly challenging problem of estimating the layout of a complex urban driving scenario. Given a single color image captured from a driving platform, we aim to predict the bird's-eye view layout of the road and other traffic participants. The estimated layout should reason beyond what is visible in the image, and compensate for the loss of 3D information due to projection. We dub this problem amodal scene layout estimation, which involves "hallucinating" scene layout for even parts of the world that are occluded in the image. To this end, we present MonoLayout, a deep neural network for real-time amodal scene layout estimation from a single image. We represent scene layout as a multi-channel semantic occupancy grid, and leverage adversarial feature learning to hallucinate plausible completions for occluded image parts. Due to the lack of fair baseline methods, we extend several state-of-the-art approaches for road-layout estimation and vehicle occupancy estimation in bird's-eye view to the amodal setup for rigorous evaluation. By leveraging temporal sensor fusion to generate training labels, we significantly outperform current art over a number of datasets. On the KITTI and Argoverse datasets, we outperform all baselines by a significant margin. We also make all our annotations, and code publicly available. A video abstract of this paper is available https://www.youtube.com/watch?v=HcroGyo6yRQ .

For agents

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

get_harvested_code_for_paper("2002.08394")
get_code_for_paper("2002.08394")
have("2002.08394")

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

curl https://syntology.ai/api/ran/2002.08394.json

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

[![Syntology run record](https://syntology.ai/api/ran/2002.08394.svg)](https://syntology.ai/paper/2002.08394)

Connect an agent — have() is free.