SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1909.09115 · 2019

Self-Supervised Learning of Depth and Motion Under Photometric Inconsistency

arXiv · PDF · Open in the Atlas

Code that ran

11 of 15 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, 1 violated it and 9 have no contract verdict; 0 failures of our instrument; 4 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/1909.09115.json · badge: /api/ran/1909.09115.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
hlzz/DeepMatchVO community (classifier label)
classifier basis not recorded on the link
11 ran · 4 no recorded run
FunctionStatusWhere it lives
associate Ran
ran on a synthesized input
hlzz/DeepMatchVO/kitti_eval/pose_evaluation_utils.py
code served (permissive licence) · get_code("3b28cf14a88f409c")
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 2, 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 torch
import numpy as np

# Construct valid arguments for the associate function
# The function expects two dictionaries mapping timestamps (keys) to data (values)
# It finds matching pairs based on time offset and max difference

first_list = {
    100.5: "data_a",
    200.3: "data_b",
    300.8: "data_c"
}

second_list = {
    100.7: "data_x",
    200.5: "data_y",
    310.0: "data_z"
}

offset = 0.2  # Time offset between the two lists
max_difference = 1.5  # Maximum allowed time difference for matching

ARGS = {
    "first_list": first_list,
    "second_list": second_list,
    "offset": offset,
    "max_difference": max_difference
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, 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
complete_batch_size Ran
ran on a synthesized input
hlzz/DeepMatchVO/common_utils.py
code served (permissive licence) · get_code("4463b24ece76265c")
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 torch
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# - input_list: a list that is modified in-place and returned
# - batch_size: an integer used for modulo operation and range calculation
# The function pads input_list to make its length divisible by batch_size
# by repeating the last element.

# Create a simple input_list with some elements
input_list = [
    torch.randn(2, 4),
    torch.randn(2, 4),
    torch.randn(2, 4),
]

# batch_size that doesn't divide evenly into len(input_list)
# len(input_list) = 3, so batch_size = 2 will leave remainder 1
batch_size = 2

ARGS = {
    "input_list": input_list,
    "batch_size": batch_size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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
compute_ate Ran
ran on a synthesized input
hlzz/DeepMatchVO/kitti_eval/pose_evaluation_utils.py
code served (permissive licence) · get_code("63782d8b81c25312")
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 float64, ndim 0, 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 temporary files with trajectory data
temp_dir = tempfile.mkdtemp()

# Ground truth file with timestamps and 3D positions (and some extra data)
gtruth_file = os.path.join(temp_dir, "gtruth.txt")
with open(gtruth_file, 'w') as f:
    f.write("1.0 0.0 0.0 0.0 0.1 0.2 0.3\n")
    f.write("1.005 0.1 0.1 0.1 0.2 0.3 0.4\n")
    f.write("1.01 0.2 0.2 0.2 0.3 0.4 0.5\n")

# Prediction file with timestamps and 3D positions (and some extra data)
pred_file = os.path.join(temp_dir, "pred.txt")
with open(pred_file, 'w') as f:
    f.write("1.0 0.05 0.05 0.05 0.1 0.2 0.3\n")
    f.write("1.005 0.15 0.15 0.15 0.2 0.3 0.4\n")
    f.write("1.01 0.25 0.25 0.25 0.3 0.4 0.5\n")

ARGS = {
    "gtruth_file": gtruth_file,
    "pred_file": pred_file
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
compute_errors Ran
ran, violated the contract we drafted
hlzz/DeepMatchVO/kitti_eval/depth_evaluation_utils.py
code served (permissive licence) · get_code("d83582de454a0b40")
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
import torch

torch.manual_seed(0)

# Create small numpy arrays for ground truth and predictions
# These should be 1D arrays of depth values (common for depth estimation tasks)
gt = np.array([1.0, 2.0, 3.0, 4.0, 5.0], dtype=np.float32)
pred = np.array([1.1, 1.9, 3.2, 3.8, 5.1], dtype=np.float32)

ARGS = {
    "gt": gt,
    "pred": pred,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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
euler2mat Ran
ran, honoured the contract we drafted
hlzz/DeepMatchVO/geo_utils.py
code served (permissive licence) · get_code("5fe6d6d6afa1a77c")
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 EagerTensor, ndim 4, dtype family float, shape (2, 1, 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
import tensorflow as tf

# Set random seed for reproducibility
np.random.seed(0)
tf.random.set_seed(0)

# Create small test tensors with shape [B, N] as specified in docstring
# B = batch size = 2, N = 1 (as per the function's internal N=1)
B = 2
N = 1

# Create euler angle inputs with shape [B, N]
z = tf.constant([[0.5], [1.0]], dtype=tf.float32)  # shape [2, 1]
y = tf.constant([[0.3], [0.7]], dtype=tf.float32)  # shape [2, 1]
x = tf.constant([[0.2], [0.4]], dtype=tf.float32)  # shape [2, 1]

ARGS = {
    'z': z,
    'y': y,
    'x': x,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 4, dtype family float, shape (2, 1, 3, 3) — shape (2, 1, 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
get_inlier_image_coords Ran
ran on a synthesized input
hlzz/DeepMatchVO/preprocess_matches.py
code served (permissive licence) · get_code("6e5b7b2f8aab8f89")
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 SIFT keys for first image
# Each key is (x, y, color, scale, orientation)
sift_keys1 = [
    (10.5, 20.3, 128, 1.2, 0.5),
    (50.2, 60.1, 64, 1.5, 1.2),
    (100.0, 150.5, 200, 2.0, 2.1),
    (200.3, 250.8, 100, 1.8, 0.9),
]

# Create SIFT keys for second image
sift_keys2 = [
    (12.1, 22.5, 130, 1.3, 0.6),
    (51.0, 61.2, 65, 1.6, 1.3),
    (102.3, 152.1, 202, 2.1, 2.2),
    (202.1, 252.3, 102, 1.9, 1.0),
]

# Create feature matches as numpy array with shape (num_matches, 3)
# Format: [index_in_keys1, index_in_keys2, flag]
# flag=2 for fundamental matrix inliers, flag=1 for homography inliers, flag=3 for both
feature_matches = np.array([
    [0, 0, 2],  # match between key 0 and 0, fundamental matrix inlier
    [1, 1, 1],  # match between key 1 and 1, homography inlier
    [2, 2, 3],  # match between key 2 and 2, both types
    [3, 3, 0],  # match between key 3 and 3, not an inlier
], dtype=np.int32)

ARGS = {
    'sift_keys1': sift_keys1,
    'sift_keys2': sift_keys2,
    'feature_matches': feature_matches,
    'type': 'f'
}
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
is_valid_sample Ran
ran on a synthesized input
hlzz/DeepMatchVO/common_utils.py
code served (permissive licence) · get_code("0e11bed3f0ee6730")
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 bool, 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 torch

# Analyze the function:
# - frames: a list that supports len() and indexing, where each element is a string with ' ' separator
# - tgt_idx: an integer index into frames
# - seq_length: an integer that determines the offset range

# Create frames as a list of strings in format "drive_id other_info"
# We need frames where all elements have the same drive_id to make is_valid_sample return True
frames = [
    "drive_001 frame_0",
    "drive_001 frame_1",
    "drive_001 frame_2",
    "drive_001 frame_3",
    "drive_001 frame_4",
]

# tgt_idx should be in the middle range so that min_src_idx and max_src_idx are valid
tgt_idx = 2

# seq_length should be chosen so that the indices stay within bounds
# max_src_offset = int((seq_length - 1) / 2)
# For tgt_idx = 2 and len(frames) = 5:
#   min_src_idx = tgt_idx - max_src_offset >= 0
#   max_src_idx = tgt_idx + max_src_offset < 5
# If seq_length = 5: max_src_offset = 2, min_src_idx = 0, max_src_idx = 4 (valid)
seq_length = 5

ARGS = {
    "frames": frames,
    "tgt_idx": tgt_idx,
    "seq_length": seq_length,
}
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 bool, 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
load_gt_disp_kitti Ran
ran on a synthesized input
hlzz/DeepMatchVO/kitti_eval/depth_evaluation_utils.py
code served (permissive licence) · get_code("6f8a54ff22b913e2")
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 numpy as np
import cv2
import tempfile
import os

# Create a temporary directory structure to simulate the KITTI dataset
temp_dir = tempfile.mkdtemp()
training_dir = os.path.join(temp_dir, 'training', 'disp_noc_0')
os.makedirs(training_dir, exist_ok=True)

# Create 200 dummy disparity images
for i in range(200):
    # Create a small dummy disparity image (uint16 format, as KITTI uses)
    # The function reads with cv2.imread(..., -1) which preserves the original format
    disp_data = np.random.randint(0, 5000, size=(32, 64), dtype=np.uint16)
    filename = os.path.join(training_dir, str(i).zfill(6) + '_10.png')
    cv2.imwrite(filename, disp_data)

ARGS = {
    'path': temp_dir
}
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 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
mat2euler Ran
ran on a synthesized input
hlzz/DeepMatchVO/geo_utils.py
code served (permissive licence) · get_code("d692f8d01af4e245")
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 SymbolicTensor. 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 tensorflow as tf
import numpy as np

# Create a valid rotation matrix batch
# Shape: [batch_size, 3, 3]
# We'll create 2 valid rotation matrices

# Identity rotation matrix
identity_rot = np.eye(3, dtype=np.float32)

# Another valid rotation (90 degree rotation around z-axis)
angle = np.pi / 2
rot_z = np.array([
    [np.cos(angle), -np.sin(angle), 0],
    [np.sin(angle), np.cos(angle), 0],
    [0, 0, 1]
], dtype=np.float32)

# Stack into batch of 2
rot_batch = np.stack([identity_rot, rot_z], axis=0)

# Convert to TensorFlow tensor
rot_tensor = tf.constant(rot_batch, dtype=tf.float32)

ARGS = {
    "rot": rot_tensor
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SymbolicTensor — a SymbolicTensor output has no array shape of its own; none is 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
pose_vec2rt Ran
ran on a synthesized input
hlzz/DeepMatchVO/geo_utils.py
code served (permissive licence) · get_code("b7fe00436895066d")
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. 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 tensorflow as tf
import numpy as np

# Set seed for reproducibility
tf.random.set_seed(0)
np.random.seed(0)

# Create a small batch of 6DoF pose vectors
# Shape: [B, 6] where B=2, and 6 = [rz, ry, rx, tx, ty, tz]
batch_size = 2
pose_vector = tf.constant([
    [0.1, 0.2, 0.3, 0.5, 0.6, 0.7],
    [0.15, 0.25, 0.35, 0.55, 0.65, 0.75]
], dtype=tf.float32)

ARGS = {
    "vec": pose_vector
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
read_file_list Ran
ran on a synthesized input
hlzz/DeepMatchVO/kitti_eval/pose_evaluation_utils.py
code served (permissive licence) · get_code("04e59d5029ab09d9")
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 trajectory data in the expected format
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_file.write("1.0 0.5 1.2 2.3 3.4\n")
temp_file.write("2.5 1.1 2.2 3.3 4.4\n")
temp_file.write("# This is a comment\n")
temp_file.write("3.7 2.0 3.1 4.2 5.3\n")
temp_file.close()

filename = temp_file.name

ARGS = {
    'filename': 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 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
convert_disps_to_depths_kitti Not yet run
unverified (harvested, no recorded run)
hlzz/DeepMatchVO/kitti_eval/depth_evaluation_utils.py
code served (permissive licence) · get_code("cf98eb3840092327")
How it ran

no invocation recorded.

read_feature_repo Not yet run
unverified (harvested, no recorded run)
hlzz/DeepMatchVO/preprocess_matches.py
code served (permissive licence) · get_code("e14a64cfd9ea52ab")
How it ran

no invocation recorded.

read_match_repo Not yet run
unverified (harvested, no recorded run)
hlzz/DeepMatchVO/preprocess_matches.py
code served (permissive licence) · get_code("e4858ffc45b28f0d")
How it ran

no invocation recorded.

resize_like Not yet run
unverified (harvested, no recorded run)
hlzz/DeepMatchVO/nets.py
code served (permissive licence) · get_code("a15c2c0e0aedaa22")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 1 the same month, 1 later) and 10 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
compute_errors 24 other papers
3 dated earlier, 1 the same month, 20 later
1609.03677 (2016-09)
Unsupervised Monocular Depth Estimation with Left-Right Consistency
1803.02276 (2018-03)
GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose
1806.01260 (2018-06)
Digging Into Self-Supervised Monocular Depth Estimation
and 21 more not named here
convert_disps_to_depths_kitti 2 other papers
2 dated earlier
1609.03677 (2016-09)
Unsupervised Monocular Depth Estimation with Left-Right Consistency
1803.02276 (2018-03)
GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose
load_gt_disp_kitti 2 other papers
2 dated earlier
1609.03677 (2016-09)
Unsupervised Monocular Depth Estimation with Left-Right Consistency
1803.02276 (2018-03)
GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose
euler2mat 1 other paper
1 dated earlier
1803.02276 (2018-03)
GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose
resize_like 1 other paper
1 dated earlier
1810.06125 (2018-10)
Every Pixel Counts ++: Joint Learning of Geometry and Motion with 3D Holistic Understanding

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 self-supervised learning of depth and pose from monocular sequences provides an attractive solution by using the photometric consistency of nearby frames as it depends much less on the ground-truth data. In this paper, we address the issue when previous assumptions of the self-supervised approaches are violated due to the dynamic nature of real-world scenes. Different from handling the noise as uncertainty, our key idea is to incorporate more robust geometric quantities and enforce internal consistency in the temporal image sequence. As demonstrated on commonly used benchmark datasets, the proposed method substantially improves the state-of-the-art methods on both depth and relative pose estimation for monocular image sequences, without adding inference overhead.

For agents

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

get_harvested_code_for_paper("1909.09115")
get_code_for_paper("1909.09115")
have("1909.09115")

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

curl https://syntology.ai/api/ran/1909.09115.json

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

[![Syntology run record](https://syntology.ai/api/ran/1909.09115.svg)](https://syntology.ai/paper/1909.09115)

Connect an agent — have() is free.