SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2312.02145 · 2023

Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation

arXiv · PDF · Open in the Atlas

Code that ran

22 of 26 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, 3 violated it and 8 have no contract verdict; 10 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/2312.02145.json · badge: /api/ran/2312.02145.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
prs-eth/marigold canonical (classifier label)
classifier basis: census
9 ran · 2 no recorded run
Magicboomliu/Accelerator-Simple-Template community (classifier label)
classifier basis not recorded on the link
9 ran · 2 no recorded run
damaggu/tadp community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
FunctionStatusWhere it lives
Disparity_Normalization Ran
ran on a synthesized input
Magicboomliu/Accelerator-Simple-Template/playground/check_depth_est.py
code served (permissive licence) · get_code("babc9c3037a5b587")
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 disparity tensor with distinct dimensions
# Shape: (2, 4) - batch size 2, 4 disparity values
disparity = torch.randn(2, 4)

ARGS = {
    "disparity": disparity
}
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
chunk Ran
ran; our contract draft was wrong, not the code
damaggu/tadp/TADP/tadp_seg.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from vicgalle/stable-diffusion-aesthetic-gradients, is pointer only (licence: NOASSERTION) · get_code("8241c0562bc710fd")
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 vicgalle/stable-diffusion-aesthetic-gradients, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

chw2hwc Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
prs-eth/marigold/marigold/util/image_util.py
code served (permissive licence) · get_code("1a9761d7649b539b")
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 3, dtype family float, shape (4, 5, 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
import numpy as np

torch.manual_seed(0)

# Test with torch.Tensor - CHW format (Channels, Height, Width)
chw_tensor = torch.randn(3, 4, 5)  # 3 channels, 4 height, 5 width

ARGS = {
    "chw": chw_tensor
}
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 Tensor, ndim 3, dtype family float, shape (4, 5, 3) — shape (4, 5, 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
colorize_depth_maps Ran
ran; our contract draft was wrong, not the code
prs-eth/marigold/marigold/util/image_util.py
code served (permissive licence) · get_code("1e519d3cea1f46e3")
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, 4, 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
import numpy as np
import matplotlib

torch.manual_seed(0)

# Create a small depth map tensor (batch=2, height=4, width=3)
depth_map_torch = torch.randn(2, 4, 3) * 5 + 10

# Define min and max depth values
min_depth = 5.0
max_depth = 15.0

# Create a valid mask (same spatial dims as depth map)
valid_mask = torch.ones(2, 4, 3, dtype=torch.bool)
valid_mask[0, 0, 0] = False  # Mark one pixel as invalid

ARGS = {
    'depth_map': depth_map_torch,
    'min_depth': min_depth,
    'max_depth': max_depth,
    'cmap': 'Spectral',
    'valid_mask': valid_mask
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy, 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, 4, 3) — shape (2, 3, 4, 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
default Ran
ran, violated the contract we drafted
damaggu/tadp/TADP/vpd/models.py
code served (permissive licence) · get_code("424012cb37b31172")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) 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.

from inspect import isfunction

def exists(val):
    return val is not None

def default(val, d):
    if exists(val):
        return val
    return d() if isfunction(d) else d

# Test case 1: val exists (not None), d is a regular value
ARGS_1 = {
    'val': 42,
    'd': 100
}

# Test case 2: val is None, d is a regular value
ARGS_2 = {
    'val': None,
    'd': 100
}

# Test case 3: val is None, d is a function
def default_func():
    return 200

ARGS_3 = {
    'val': None,
    'd': default_func
}

# Test case 4: val exists, d is a function
ARGS_4 = {
    'val': "existing_value",
    'd': lambda: "default_value"
}

# Use one of the test cases as the final ARGS
ARGS = ARGS_3
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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 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
depth2disp Ran
ran on a synthesized input
Magicboomliu/Accelerator-Simple-Template/dataloader/file_io.py
code served (permissive licence) · get_code("7d50f6c8900b3933")
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
from PIL import Image
import tempfile
import os

# Create a temporary depth image file
# The function opens an image and converts it to a numpy array
# Depth values are divided by 10000, so we create an image with reasonable depth values
temp_dir = tempfile.mkdtemp()
depth_image_path = os.path.join(temp_dir, "depth.png")

# Create a small depth image (16-bit grayscale would be typical, but PIL Image.open returns what we save)
# Create a simple numpy array and save it as a PNG
depth_array = np.array([[5000, 6000], [7000, 8000]], dtype=np.uint16)
depth_img = Image.fromarray(depth_array, mode='I;16')
depth_img.save(depth_image_path)

ARGS = {
    "left_depth_path": depth_image_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, 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
depth2disp_cv Ran
ran on a synthesized input
Magicboomliu/Accelerator-Simple-Template/dataloader/file_io.py
code served (permissive licence) · get_code("12e95408bfe2ed81")
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 cv2
import numpy as np
import tempfile
import os

# Create a temporary directory for test files
temp_dir = tempfile.mkdtemp()

# Create a small test depth image
# The function expects cv2.imread to read an image file
# We'll create a simple depth map and save it
depth_map = np.array([[1000, 2000], [3000, 4000]], dtype=np.uint16)
depth_path = os.path.join(temp_dir, "test_depth.png")
cv2.imwrite(depth_path, depth_map)

ARGS = {
    "left_depth_path": depth_path
}
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
ensemble_depth Ran
ran on a synthesized input
prs-eth/marigold/marigold/util/ensemble.py
code served (permissive licence) · get_code("55702399683243f2")
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 torch
import numpy as np

torch.manual_seed(0)

# Create a small ensemble depth tensor with shape (B, 1, H, W)
# B=2 ensemble members, H=4, W=3
depth = torch.randn(2, 1, 4, 3).abs()  # depth values should be non-negative

ARGS = {
    'depth': depth,
    'scale_invariant': True,
    'shift_invariant': True,
    'output_uncertainty': False,
    'reduction': 'median',
    'regularizer_strength': 0.02,
    'max_iter': 50,
    'tol': 1e-06,
    'max_res': 1024,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, torch — 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
ensemble_iid Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
prs-eth/marigold/marigold/util/ensemble.py
code served (permissive licence) · get_code("8dd1994ceb1eb671")
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 4, 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

torch.manual_seed(0)

# Create a small tensor for ensemble predictions
# Shape: (num_models, batch_size, output_dim)
# Using distinct dimensions: 3 models, 2 batch samples, 4 output dimensions
targets = torch.randn(3, 2, 4)

ARGS = {
    'targets': targets,
    'output_uncertainty': True,
    'reduction': 'median'
}
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 tuple, ndim 4, 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
ensemble_normals Ran
ran; our contract draft was wrong, not the code
prs-eth/marigold/marigold/util/ensemble.py
code served (permissive licence) · get_code("6f901ea1ce90a56a")
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 torch
import numpy as np

torch.manual_seed(0)

# Create a small ensemble of normal maps
# Shape: (B, 3, H, W) where B is number of ensemble members
B = 2  # 2 ensemble members
C = 3  # 3 channels (x, y, z normal components)
H = 4  # height
W = 5  # width

# Create random normals and normalize them to unit vectors
normals = torch.randn(B, C, H, W)
# Normalize each normal vector
norm = torch.norm(normals, dim=1, keepdim=True)
normals = normals / norm.clamp(min=1e-6)

ARGS = {
    'normals': normals,
    'output_uncertainty': False,
    'reduction': 'closest'
}
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 — 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
exists Ran
ran, violated the contract we drafted
damaggu/tadp/TADP/vpd/models.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ThomasMrY/VCT, is pointer only (licence: NONE) · get_code("aa5486a3650902d8")
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 ThomasMrY/VCT, 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.

extract_into_tensor Ran
ran; our contract draft was wrong, not the code
prs-eth/marigold/src/trainer/marigold_depth_LCM_trainer.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from Luo-Yihong/YOSO, is pointer only (licence: NONE) · get_code("a15cfd7932844ef9")
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 Luo-Yihong/YOSO, 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_predicted_original_sample Ran
ran; our contract draft was wrong, not the code
prs-eth/marigold/src/trainer/marigold_depth_LCM_trainer.py
code served (permissive licence) · get_code("1950491589be0476")
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.nn as nn

torch.manual_seed(0)

# Create small tensors for the function
batch_size = 2
sample_dim = 4
num_timesteps = 10

# model_output: tensor with shape (batch_size, sample_dim)
model_output = torch.randn(batch_size, sample_dim)

# timesteps: tensor with shape (batch_size,) containing indices
timesteps = torch.randint(0, num_timesteps, (batch_size,))

# sample: tensor with shape (batch_size, sample_dim)
sample = torch.randn(batch_size, sample_dim)

# prediction_type: string, one of 'epsilon', 'sample', 'v_prediction'
prediction_type = 'epsilon'

# alphas: tensor with shape (num_timesteps,) for gathering
alphas = torch.linspace(0.1, 1.0, num_timesteps)

# sigmas: tensor with shape (num_timesteps,) for gathering
sigmas = torch.linspace(0.9, 0.1, num_timesteps)

ARGS = {
    'model_output': model_output,
    'timesteps': timesteps,
    'sample': sample,
    'prediction_type': prediction_type,
    'alphas': alphas,
    'sigmas': sigmas,
}
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
int_list Ran
ran, honoured the contract we drafted
Magicboomliu/Accelerator-Simple-Template/dataloader/utils.py
code served (permissive licence) · get_code("76f433ffb0eb1150")
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 int; 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

def int_list(s):
    """Convert string to int list"""
    return [int(x) for x in s.split(',')]

ARGS = {
    "s": "1,2,3,4,5"
}
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 list, ndim 1, dtype family int — 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
kitti_colormap Ran
ran on a synthesized input
Magicboomliu/Accelerator-Simple-Template/utils/colormap.py
code served (permissive licence) · get_code("50a3e2b7e8148db0")
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 (4, 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

# Create a small disparity map (HxW)
# Using a 4x3 array to have distinct height and width
disparity = np.random.rand(4, 3).astype(np.float32) * 50.0

ARGS = {
    "disparity": disparity,
    "maxval": -1
}
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 3, dtype family int, shape (4, 3, 3) — shape (4, 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
read_16bit_gt Ran
ran on a synthesized input
Magicboomliu/Accelerator-Simple-Template/utils/colormap.py
code served (permissive licence) · get_code("bf3d969e975a0237")
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 (4, 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
import tempfile
import os

# Create a temporary 16-bit image file for testing
temp_dir = tempfile.mkdtemp()
temp_image_path = os.path.join(temp_dir, "test_16bit.png")

# Create a simple 16-bit grayscale image (4x3 for distinct dimensions)
image_data = np.array([
    [256, 512, 768],
    [1024, 1280, 1536],
    [1792, 2048, 2304],
    [2560, 2816, 3072]
], dtype=np.uint16)

# Write as 16-bit PNG
cv2.imwrite(temp_image_path, image_data)

ARGS = {
    "path": temp_image_path
}
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 (4, 3) — shape (4, 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_img Ran
ran on a synthesized input
Magicboomliu/Accelerator-Simple-Template/dataloader/file_io.py
code served (permissive licence) · get_code("a5edaeb9db495d4c")
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 float, shape (4, 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
from PIL import Image
import tempfile
import os

# Create a temporary RGB image file
temp_dir = tempfile.mkdtemp()
temp_image_path = os.path.join(temp_dir, "test_image.png")

# Create a small RGB image (4x3 to have distinct dimensions)
img_array = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)
img = Image.fromarray(img_array, mode='RGB')
img.save(temp_image_path)

ARGS = {
    "filename": temp_image_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, 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 float, shape (4, 3, 3) — shape (4, 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
read_text_lines Ran
ran; our contract draft was wrong, not the code
Magicboomliu/Accelerator-Simple-Template/dataloader/utils.py
code served (permissive licence) · get_code("78088e741a0c4b80")
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 text file with some content
temp_dir = tempfile.mkdtemp()
filepath = os.path.join(temp_dir, "test.txt")

with open(filepath, 'w') as f:
    f.write("line 1\n")
    f.write("line 2  \n")
    f.write("line 3\n")

ARGS = {
    "filepath": filepath
}
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
resize_max_res Ran
ran; our contract draft was wrong, not the code
prs-eth/marigold/marigold/util/image_util.py
code served (permissive licence) · get_code("715b48a5e7426c5e")
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, 4, 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
from torchvision.transforms import InterpolationMode

torch.manual_seed(0)

# Create a small batch of images with shape [B, C, H, W]
# B=2 (batch), C=3 (RGB), H=8, W=6
img = torch.randint(0, 256, (2, 3, 8, 6), dtype=torch.uint8).float() / 255.0

# Maximum edge resolution
max_edge_resolution = 4

# Resampling method
resample_method = InterpolationMode.BILINEAR

ARGS = {
    "img": img,
    "max_edge_resolution": max_edge_resolution,
    "resample_method": resample_method,
}
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 Tensor, ndim 4, dtype family float, shape (2, 3, 4, 3) — shape (2, 3, 4, 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
resize_max_res_tensor Ran
ran on a synthesized input
Magicboomliu/Accelerator-Simple-Template/playground/check_depth_est.py
code served (permissive licence) · get_code("ff6b98fbbe52720a")
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, 614, 768) 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 small input tensor with shape (batch=2, channels=3, height=4, width=5)
input_tensor = torch.randn(2, 3, 4, 5)

# Define ARGS dict with all parameters
ARGS = {
    'input_tensor': input_tensor,
    'is_disp': False,
    'recom_resolution': 768
}
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, 614, 768) — shape (2, 3, 614, 768) 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
scalings_for_boundary_conditions Ran
ran, violated the contract we drafted
prs-eth/marigold/src/trainer/marigold_depth_LCM_trainer.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from layer6ai-labs/direct-cms, is pointer only (licence: NONE) · get_code("74a9b4e5e34c0860")
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 layer6ai-labs/direct-cms, 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.

uniq Ran
ran; our contract draft was wrong, not the code
damaggu/tadp/TADP/vpd/models.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from JingWu321/MUNBa, is pointer only (licence: NONE) · get_code("9a299fe5ae09e407")
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 JingWu321/MUNBa, 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.

find_batch_size Not yet run
unverified (harvested, no recorded run)
prs-eth/marigold/marigold/util/batchsize.py
code served (permissive licence) · get_code("55c27917099f7dc3")
How it ran

no invocation recorded.

find_batch_size Not yet run
unverified (harvested, no recorded run)
Magicboomliu/Accelerator-Simple-Template/utils/batch_size.py
code served (permissive licence) · get_code("28a8af1b1cd8cb05")
How it ran

no invocation recorded.

find_occ_mask Not yet run
unverified (harvested, no recorded run)
Magicboomliu/Accelerator-Simple-Template/dataloader/utils.py
code served (permissive licence) · get_code("bd107ff2d1229ab5")
How it ran

no invocation recorded.

get_pred_name Not yet run
unverified (harvested, no recorded run)
prs-eth/marigold/src/dataset/base_depth_dataset.py
code served (permissive licence) · get_code("19b4e7dbd2c70b35")
How it ran

no invocation recorded.

Also present in other papers' code

Of 26 samples lifted from this paper's linked repositories, 26 carry a function identity and 0 do not. Of the 25 distinct functions with an identity, 11 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 5 the same month, 9 later) and 14 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
exists 486 other papers
262 dated earlier, 16 the same month, 202 later, 4 same year, order not determinable, 2 undated · 78 only as an assembly brick
1603.08155 (2016-03)
Perceptual Losses for Real-Time Style Transfer and Super-Resolution
1611.07004 (2016-11)
Image-to-Image Translation with Conditional Adversarial Networks
1804.04637 (2018-04)
EMBER: An Open Dataset for Training Static PE Malware Machine Learning Models
and 483 more not named here
default 207 other papers
100 dated earlier, 14 the same month, 91 later, 2 same year, order not determinable · 41 only as an assembly brick
1711.11585 (2017-11)
High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.08162 (2020-10)
SidechainNet: An All-Atom Protein Structure Dataset for Machine Learning
and 204 more not named here
uniq 115 other papers
45 dated earlier, 10 the same month, 58 later, 2 same year, order not determinable
2112.10752 (2021-12)
High-Resolution Image Synthesis with Latent Diffusion Models
2204.06125 (2022-04)
Hierarchical Text-Conditional Image Generation with CLIP Latents
2206.02779 (2022-06)
Blended Latent Diffusion
and 112 more not named here
chunk 40 other papers
27 dated earlier, 2 the same month, 11 later
2106.14490 (2021-06)
Making Images Real Again: A Comprehensive Survey on Deep Image Composition
2112.10752 (2021-12)
High-Resolution Image Synthesis with Latent Diffusion Models
2204.11824 (2022-04)
Semi-Parametric Neural Image Synthesis
and 37 more not named here
extract_into_tensor 32 other papers
8 dated earlier, 3 the same month, 21 later · 7 only as an assembly brick
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.02502 (2020-10)
Denoising Diffusion Implicit Models
2112.02815 (2021-12)
Make It Move: Controllable Image-to-Video Generation with Text Descriptions
and 29 more not named here
get_pred_name 2 other papers
2 later
2504.11457 (2025-04)
Aligning Generative Denoising with Discriminative Objectives Unleashes Diffusion for Visual Perception
2609.02854 (2026-09)
MuyBridge: Mobile Human Center-of-Mass Estimation from Monocular Video via Sparse Fusion
read_text_lines 2 other papers
2 dated earlier
2104.13918 (2021-04)
High-Resolution Optical Flow from 1D Attention and Correlation
2308.09383 (2023-08)
Label-Free Event-based Object Recognition via Joint Learning with Image Reconstruction from Events
colorize_depth_maps 1 other paper
1 later
2410.18538 (2024-10)
SMITE: Segment Me In TimE
int_list 1 other paper
1 dated earlier
2104.13918 (2021-04)
High-Resolution Optical Flow from 1D Attention and Correlation
resize_max_res 1 other paper
1 later
2409.18124 (2024-09)
Lotus: Diffusion-based Visual Foundation Model for High-quality Dense Prediction
scalings_for_boundary_conditions 1 other paper
1 later
2411.08954 (2024-11)
Inconsistencies In Consistency Models: Better ODE Solving Does Not Imply Better Samples

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

Monocular depth estimation is a fundamental computer vision task. Recovering 3D depth from a single image is geometrically ill-posed and requires scene understanding, so it is not surprising that the rise of deep learning has led to a breakthrough. The impressive progress of monocular depth estimators has mirrored the growth in model capacity, from relatively modest CNNs to large Transformer architectures. Still, monocular depth estimators tend to struggle when presented with images with unfamiliar content and layout, since their knowledge of the visual world is restricted by the data seen during training, and challenged by zero-shot generalization to new domains. This motivates us to explore whether the extensive priors captured in recent generative diffusion models can enable better, more generalizable depth estimation. We introduce Marigold, a method for affine-invariant monocular depth estimation that is derived from Stable Diffusion and retains its rich prior knowledge. The estimator can be fine-tuned in a couple of days on a single GPU using only synthetic training data. It delivers state-of-the-art performance across a wide range of datasets, including over 20% performance gains in specific cases. Project page: https://marigoldmonodepth.github.io.

For agents

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

get_harvested_code_for_paper("2312.02145")
get_code_for_paper("2312.02145")
have("2312.02145")

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

curl https://syntology.ai/api/ran/2312.02145.json

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

[![Syntology run record](https://syntology.ai/api/ran/2312.02145.svg)](https://syntology.ai/paper/2312.02145)

Connect an agent — have() is free.