SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.08854 · 2020

DVI: Depth Guided Video Inpainting for Autonomous Driving

arXiv · PDF · Open in the Atlas

Code that ran

10 of 10 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; 3 failures of our instrument; 0 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2007.08854.json · badge: /api/ran/2007.08854.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
ApolloScapeAuto/dataset-api canonical (classifier label)
classifier basis: census
10 ran · 0 no recorded run
FunctionStatusWhere it lives
to_np Ran
ran; our contract draft was wrong, not the code
ApolloScapeAuto/dataset-api/stereo/finetune.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from deeplearning-wisc/hypo, is pointer only (licence: NONE) · get_code("8c6b999f5a291101")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from deeplearning-wisc/hypo, 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.

IOU Ran
ran on a synthesized input
ApolloScapeAuto/dataset-api/utils/eval_utils.py
code served (permissive licence) · get_code("01ccff28924c5c5f")
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 float32, ndim 0, dtype family float, 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 numpy as np

# Create two simple binary masks with some overlap
mask1 = np.array([
    [1, 1, 0, 0],
    [1, 1, 0, 0],
    [0, 0, 0, 0],
    [0, 0, 0, 0]
], dtype=np.float32)

mask2 = np.array([
    [1, 0, 0, 0],
    [0, 1, 1, 0],
    [0, 1, 1, 0],
    [0, 0, 0, 0]
], dtype=np.float32)

ARGS = {
    "mask1": mask1,
    "mask2": mask2
}
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 float32, ndim 0, dtype family float, 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
euler_angles_to_quaternions Ran
ran on a synthesized input
ApolloScapeAuto/dataset-api/utils/utils.py
code served (permissive licence) · get_code("44d2bc18dfc66e50")
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, 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

# Create a small n x 3 matrix of euler angles (roll, pitch, yaw)
# Using 2 samples with 3 angles each for minimal but meaningful test
angle = np.array([
    [0.1, 0.2, 0.3],
    [0.4, 0.5, 0.6]
], dtype=np.float32)

ARGS = {
    "angle": angle
}
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 (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
evaluation Ran
ran on a synthesized input
ApolloScapeAuto/dataset-api/trajectory_prediction/evaluation.py
code served (permissive licence) · get_code("45585870c1fd0413")
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 tuple, 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

# Understand the function: it processes frame data with predictions and ground truth
# Each frame contains objects with: [id, type, x_pos, y_pos, ...]
# type: 1,2=vehicle, 3=pedestrian, 4=bicycle, 5=skip
# consider_peds is a list of lists of object IDs to consider per sequence

# Create sample frame data for predictions
# Each frame is a list of [id, type, x, y]
frame_data_result = [
    # Frame 0-5 (sequence 1)
    [[1, 1, 0.0, 0.0], [2, 3, 5.0, 5.0]],  # frame 0
    [[1, 1, 0.5, 0.5], [2, 3, 5.2, 5.1]],  # frame 1
    [[1, 1, 1.0, 1.0], [2, 3, 5.4, 5.2]],  # frame 2
    [[1, 1, 1.5, 1.5], [2, 3, 5.6, 5.3]],  # frame 3
    [[1, 1, 2.0, 2.0], [2, 3, 5.8, 5.4]],  # frame 4
    [[1, 1, 2.5, 2.5], [2, 3, 6.0, 5.5]],  # frame 5
    # Frame 6-11 (sequence 2)
    [[3, 4, 10.0, 10.0], [4, 2, 20.0, 20.0]],  # frame 6
    [[3, 4, 10.3, 10.3], [4, 2, 20.5, 20.5]],  # frame 7
    [[3, 4, 10.6, 10.6], [4, 2, 21.0, 21.0]],  # frame 8
    [[3, 4, 10.9, 10.9], [4, 2, 21.5, 21.5]],  # frame 9
    [[3, 4, 11.2, 11.2], [4, 2, 22.0, 22.0]],  # frame 10
    [[3, 4, 11.5, 11.5], [4, 2, 22.5, 22.5]],  # frame 11
]

# Create ground truth frame data
frame_data_gt = [
    # Frame 0-5 (sequence 1)
    [[1, 1, 0.0, 0.0], [2, 3, 5.0, 5.0]],  # frame 0
    [[1, 1, 0.5, 0.5], [2, 3, 5.2, 5.1]],  # frame 1
    [[1, 1, 1.0, 1.0], [2, 3, 5.4, 5.2]],  # frame 2
    [[1, 1, 1.5, 1.5], [2, 3, 5.6, 5.3]],  # frame 3
    [[1, 1, 2.0, 2.0], [2, 3, 5.8, 5.4]],  # frame 4
    [[1, 1, 2.5, 2.5], [2, 3, 6.0, 5.5]],  # frame 5
    # Frame 6-11 (sequence 2)
    [[3, 4, 10.0, 10.0], [4, 2, 20.0, 20.0]],  # frame 6
    [[3, 4, 10.3, 10.3], [4, 2, 20.5, 20.5]],  # frame 7
    [[3, 4, 10.6, 10.6], [4, 2, 21.0, 21.0]],  # frame 8
    [[3, 4, 10.9, 10.9], [4, 2, 21.5, 21.5]],  # frame 9
    [[3, 4, 11.2, 11.2], [4, 2, 22.0, 22.0]],  # frame 10
    [[3, 4, 11.5, 11.5], [4, 2, 22.5, 22.5]],  # frame 11
]

# consider_peds: list of lists of IDs to consider per sequence
# 2 sequences of 6 frames each, so 2 groups
consider_peds = [
    [1, 2],  # sequence 1: consider IDs 1 and 2
    [3, 4],  # sequence 2: consider IDs 3 and 4
]

ARGS = {
    'frame_data_result': frame_data_result,
    'frame_data_gt': frame_data_gt,
    'consider_peds': consider_peds,
}
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 tuple, ndim 1, dtype family float — a tuple 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
intrinsic_mat_to_vec Ran
ran on a synthesized input
ApolloScapeAuto/dataset-api/utils/utils.py
code served (permissive licence) · get_code("8d7b4ed5ab9095e6")
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

# Create a 3x3 intrinsic matrix K with camera parameters
# Format: [[fx,  0, cx],
#          [ 0, fy, cy],
#          [ 0,  0,  1]]
K = np.array([
    [500.0, 0.0, 320.0],
    [0.0, 500.0, 240.0],
    [0.0, 0.0, 1.0]
], dtype=np.float32)

ARGS = {
    "K": K
}
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
intrinsic_vec_to_mat Ran
ran on a synthesized input
ApolloScapeAuto/dataset-api/utils/utils.py
code served (permissive licence) · get_code("92b17e8bdc17f325")
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 (3, 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

# The function expects:
# - intrinsic: a 4-element array-like with focal lengths and principal point offsets
# - shape: optional list/tuple of [height, width] dimensions

# Create a small intrinsic vector (4 elements: fx, fy, cx, cy normalized)
intrinsic = np.array([0.5, 0.6, 0.25, 0.3], dtype=np.float32)

# Create a shape tuple [height, width]
shape = [480, 640]

ARGS = {
    "intrinsic": intrinsic,
    "shape": shape
}
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 (3, 3) — shape (3, 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
readConsiderObjects Ran
ran on a synthesized input
ApolloScapeAuto/dataset-api/trajectory_prediction/evaluation.py
code served (permissive licence) · get_code("c16fa880c6501e54")
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 tempfile
import os

# Create a temporary file with test data
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_file.write("1 2 3 4\n")
temp_file.write("5 6 7 8\n")
temp_file.write("9 10 11 12\n")
temp_file.close()

ARGS = {
    'filename': temp_file.name
}
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 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
readTrajectory Ran
ran on a synthesized input
ApolloScapeAuto/dataset-api/trajectory_prediction/evaluation.py
code served (permissive licence) · get_code("6cbedf12c3e36ab5")
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 3, 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 tempfile
import os

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

# Write sample trajectory data: timestamp, id, type, x, y
trajectory_data = """0.0 1 0 10.5 20.3
0.0 2 1 15.2 25.1
1.0 1 0 11.0 21.0
1.0 2 1 15.8 25.5
2.0 1 0 11.5 21.5
2.0 3 0 30.0 40.0
"""

temp_file.write(trajectory_data)
temp_file.close()

ARGS = {
    'filename': temp_filename
}
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 list, ndim 3, 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
to_var Ran
ran; our contract draft was wrong, not the code
ApolloScapeAuto/dataset-api/stereo/finetune.py
code served (permissive licence) · get_code("ec6d05f0901dc0b8")
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. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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
from torch.autograd import Variable

torch.manual_seed(0)

# The function to_var takes a single parameter 'x' which should be a tensor
# It optionally moves it to CUDA (if available) and wraps it in Variable
# We need to provide a small tensor as input

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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
update_model Ran
ran; our contract draft was wrong, not the code
ApolloScapeAuto/dataset-api/stereo/finetune.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from htx0601/AGG-Net, is pointer only (licence: NONE) · get_code("885f7d8c5df224ca")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from htx0601/AGG-Net, 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.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 3 later) and 7 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
to_np 8 other papers
3 dated earlier, 5 later
1807.07860 (2018-07)
Talking Face Generation by Adversarially Disentangled Audio-Visual Representation
1811.09720 (2018-11)
Representer Point Selection for Explaining Deep Neural Networks
1902.09513 (2019-02)
FEELVOS: Fast End-to-End Embedding Learning for Video Object Segmentation
and 5 more not named here
to_var 4 other papers
3 dated earlier, 1 later
1704.04394 (2017-04)
DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agents
1905.01354 (2019-05)
Controllable Artistic Text Style Transfer via Shape-Matching GAN
2001.02890 (2020-01)
Deep Plastic Surgery: Robust and Controllable Image Editing with Human-Drawn Sketches
and 1 more not named here
update_model 1 other paper
1 later
2309.01624 (2023-09)
AGG-Net: Attention Guided Gated-convolutional Network for Depth Image Completion

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

To get clear street-view and photo-realistic simulation in autonomous driving, we present an automatic video inpainting algorithm that can remove traffic agents from videos and synthesize missing regions with the guidance of depth/point cloud. By building a dense 3D map from stitched point clouds, frames within a video are geometrically correlated via this common 3D map. In order to fill a target inpainting area in a frame, it is straightforward to transform pixels from other frames into the current one with correct occlusion. Furthermore, we are able to fuse multiple videos through 3D point cloud registration, making it possible to inpaint a target video with multiple source videos. The motivation is to solve the long-time occlusion problem where an occluded area has never been visible in the entire video. To our knowledge, we are the first to fuse multiple videos for video inpainting. To verify the effectiveness of our approach, we build a large inpainting dataset in the real urban road environment with synchronized images and Lidar data including many challenge scenes, e.g., long time occlusion. The experimental results show that the proposed approach outperforms the state-of-the-art approaches for all the criteria, especially the RMSE (Root Mean Squared Error) has been reduced by about 13%.

For agents

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

get_harvested_code_for_paper("2007.08854")
get_code_for_paper("2007.08854")
have("2007.08854")

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

curl https://syntology.ai/api/ran/2007.08854.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.08854.svg)](https://syntology.ai/paper/2007.08854)

Connect an agent — have() is free.