SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.05233 · 2020

Continual Adaptation for Deep Stereo

arXiv · PDF · Open in the Atlas

Code that ran

8 of 16 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, 0 violated it and 7 have no contract verdict; 0 failures of our instrument; 8 no recorded run), measured 2026-11-21; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2007.05233.json · badge: /api/ran/2007.05233.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
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo canonical (classifier label)
classifier basis: census
8 ran · 8 no recorded run
FunctionStatusWhere it lives
l1 Ran
ran on a synthesized input
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Losses/loss_factory.py
code served (permissive licence) · get_code("f7bd997c454b72ef")
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 3, dtype family float, shape (2, 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 tensorflow as tf
import numpy as np

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

# Create small tensor arguments
batch_size = 2
height = 4
width = 3

# Create predicted image (x)
x = tf.constant(np.random.randn(batch_size, height, width).astype(np.float32))

# Create target image (y)
y = tf.constant(np.random.randn(batch_size, height, width).astype(np.float32))

# Create mask (optional)
mask = tf.ones((batch_size, height, width), dtype=tf.float32)

ARGS = {
    "x": x,
    "y": y,
    "mask": mask
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 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
l2 Ran
ran on a synthesized input
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Losses/loss_factory.py
code served (permissive licence) · get_code("554090648d5d1688")
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 3, dtype family float, shape (2, 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 tensorflow as tf
import numpy as np

# Create small test tensors
batch_size = 2
height = 4
width = 3

# Predicted image (x)
x = tf.constant(
    np.random.randn(batch_size, height, width).astype(np.float32),
    dtype=tf.float32
)

# Target image (y)
y = tf.constant(
    np.random.randn(batch_size, height, width).astype(np.float32),
    dtype=tf.float32
)

# Mask (optional, but test with it)
mask = tf.constant(
    np.random.randint(0, 2, size=(batch_size, height, width)).astype(np.float32),
    dtype=tf.float32
)

ARGS = {
    "x": x,
    "y": y,
    "mask": mask
}
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 3, dtype family float, shape (2, 4, 3) — shape (2, 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
mean_l1 Ran
ran on a synthesized input
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Losses/loss_factory.py
code served (permissive licence) · get_code("741b7a75bf889cab")
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 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 tensorflow as tf
import numpy as np

# Create small test tensors for predicted and target images
# Using distinct dimensions: batch=2, height=4, width=3, channels=1
x = tf.constant(np.random.randn(2, 4, 3, 1).astype(np.float32))
y = tf.constant(np.random.randn(2, 4, 3, 1).astype(np.float32))

# Create a mask with the same shape, with some zeros to test masking
mask = tf.constant(np.random.choice([0.0, 1.0], size=(2, 4, 3, 1)).astype(np.float32))

ARGS = {
    "x": x,
    "y": y,
    "mask": mask,
}
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 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
pad_image Ran
ran on a synthesized input
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Data_utils/preprocessing.py
code served (permissive licence) · get_code("683bb31446329db4")
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, 256, 256, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Create a small test image tensor with shape [batch, height, width, channels]
# Using distinct sizes: batch=2, height=100, width=150, channels=3
test_image = tf.constant(np.random.randn(2, 100, 150, 3).astype(np.float32))

ARGS = {
    "immy": test_image,
    "down_factor": 256,
    "dynamic": False,
}
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, 256, 256, 3) — shape (2, 256, 256, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
readPFM Ran
ran on a synthesized input
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Data_utils/continual_data_reader.py
code served (permissive licence) · get_code("5bd87bf6c7536635")
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 numpy as np
import tempfile
import os

# Create a temporary PFM file for testing
temp_dir = tempfile.mkdtemp()
pfm_file = os.path.join(temp_dir, "test.pfm")

# Create a simple PFM file (color image, 2x3 pixels)
with open(pfm_file, 'wb') as f:
    # Write header for color PFM
    f.write(b'PF\n')
    # Write dimensions
    f.write(b'2 3\n')
    # Write scale (positive for big-endian)
    f.write(b'1.0\n')
    # Write pixel data (2x3 color image = 18 floats)
    data = np.arange(18, dtype='>f4')
    f.write(data.tobytes())

ARGS = {
    'file': pfm_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 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_list_file Ran
ran on a synthesized input
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Data_utils/continual_data_reader.py
code served (permissive licence) · get_code("df3fd5fdcb3396f4")
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 tempfile
import os

# Create a temporary file with the expected format
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_file.write("# This is a comment line\n")
temp_file.write("left_img_1.png;right_img_1.png;gt_disp_1.png;conf_1.png\n")
temp_file.write("left_img_2.png;right_img_2.png;gt_disp_2.png;conf_2.png\n")
temp_file.write("left_img_3.png;right_img_3.png;gt_disp_3.png\n")
temp_file.write("left_img_4.png;right_img_4.png\n")
temp_file.close()

ARGS = {
    'path_file': 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 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_list_file Ran
ran on a synthesized input
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Data_utils/data_reader.py
code served (permissive licence) · get_code("9a30df72bada5055")
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 tempfile
import os

# Create a temporary file with dataset description
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_file.write("left_img1.png;right_img1.png;gt_disp1.pfm;conf1.pfm\n")
temp_file.write("left_img2.png;right_img2.png;gt_disp2.pfm;conf2.pfm\n")
temp_file.write("# This is a comment line\n")
temp_file.write("\n")
temp_file.write("left_img3.png;right_img3.png\n")
temp_file.flush()
temp_file.close()

path_file = temp_file.name

ARGS = {
    'path_file': path_file
}
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 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
softmax Ran
ran, honoured the contract we drafted
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Stereo_Continual_Adaptation.py
code served (permissive licence) · get_code("a49ec0b422e77ed3")
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) 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

# The softmax function takes a single parameter 'x' which should be a numpy array
# The function computes exp(x) and normalizes by sum along axis 0
# So x should be a 2D array to make axis=0 meaningful

x = np.array([
    [1.0, 2.0, 3.0, 4.0],
    [2.0, 3.0, 4.0, 5.0]
], dtype=np.float32)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
augment Not yet run
unverified (harvested, no recorded run)
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Data_utils/preprocessing.py
code served (permissive licence) · get_code("27cc437d9adfb80e")
How it ran

no invocation recorded.

check_for_weights_or_restore_them Not yet run
unverified (harvested, no recorded run)
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Data_utils/weights_utils.py
code served (permissive licence) · get_code("73c89e4ebc93f573")
How it ran

no invocation recorded.

correlation Not yet run
unverified (harvested, no recorded run)
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Nets/sharedLayers.py
code served (permissive licence) · get_code("e0bb29258316c3e5")
How it ran

no invocation recorded.

correlation_native Not yet run
unverified (harvested, no recorded run)
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Nets/sharedLayers.py
code served (permissive licence) · get_code("4e0ad923ad4f4da7")
How it ran

no invocation recorded.

correlation_tf Not yet run
unverified (harvested, no recorded run)
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Nets/sharedLayers.py
code served (permissive licence) · get_code("539df991fa984e18")
How it ran

no invocation recorded.

get_var_to_restore_list Not yet run
unverified (harvested, no recorded run)
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Data_utils/weights_utils.py
code served (permissive licence) · get_code("52ac8b9a9d55427b")
How it ran

no invocation recorded.

random_crop Not yet run
unverified (harvested, no recorded run)
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Data_utils/preprocessing.py
code served (permissive licence) · get_code("1b7e0b1f89402910")
How it ran

no invocation recorded.

read_image_from_disc Not yet run
unverified (harvested, no recorded run)
CVLAB-Unibo/Real-time-self-adaptive-deep-stereo/Data_utils/continual_data_reader.py
code served (permissive licence) · get_code("dd9ab5ac64bbf156")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 10 are also present in other papers whose code we hold (10 with at least one paper dated earlier, 0 the same month, 1 later) and 6 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
softmax 5 other papers
3 dated earlier, 2 later
1301.3781 (2013-01)
Efficient Estimation of Word Representations in Vector Space
1503.02531 (2015-03)
Distilling the Knowledge in a Neural Network
2004.08056 (2020-04)
Dialogue-Based Relation Extraction
and 2 more not named here
augment 1 other paper
1 dated earlier
1904.02957 (2019-04)
Learning to Adapt for Stereo
check_for_weights_or_restore_them 1 other paper
1 dated earlier
1904.02957 (2019-04)
Learning to Adapt for Stereo
get_var_to_restore_list 1 other paper
1 dated earlier
1904.02957 (2019-04)
Learning to Adapt for Stereo
l1 1 other paper
1 dated earlier
1904.02957 (2019-04)
Learning to Adapt for Stereo
l2 1 other paper
1 dated earlier
1904.02957 (2019-04)
Learning to Adapt for Stereo
mean_l1 1 other paper
1 dated earlier
1904.02957 (2019-04)
Learning to Adapt for Stereo
pad_image 1 other paper
1 dated earlier
1904.02957 (2019-04)
Learning to Adapt for Stereo
readPFM 1 other paper
1 dated earlier
1904.02957 (2019-04)
Learning to Adapt for Stereo
read_image_from_disc 1 other paper
1 dated earlier
1904.02957 (2019-04)
Learning to Adapt for Stereo

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-21.

Repositories linked to this paper

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

Abstract

Depth estimation from stereo images is carried out with unmatched results by convolutional neural networks trained end-to-end to regress dense disparities. Like for most tasks, this is possible if large amounts of labelled samples are available for training, possibly covering the whole data distribution encountered at deployment time. Being such an assumption systematically unmet in real applications, the capacity of adapting to any unseen setting becomes of paramount importance. Purposely, we propose a continual adaptation paradigm for deep stereo networks designed to deal with challenging and ever-changing environments. We design a lightweight and modular architecture, Modularly ADaptive Network (MADNet), and formulate Modular ADaptation algorithms (MAD, MAD++) which permit efficient optimization of independent sub-portions of the entire network. In our paradigm, the learning signals needed to continuously adapt models online can be sourced from self-supervision via right-to-left image warping or from traditional stereo algorithms. With both sources, no other data than the input images being gathered at deployment time are needed. Thus, our network architecture and adaptation algorithms realize the first real-time self-adaptive deep stereo system and pave the way for a new paradigm that can facilitate practical deployment of end-to-end architectures for dense disparity regression.

For agents

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

get_harvested_code_for_paper("2007.05233")
get_code_for_paper("2007.05233")
have("2007.05233")

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

curl https://syntology.ai/api/ran/2007.05233.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.05233.svg)](https://syntology.ai/paper/2007.05233)

Connect an agent — have() is free.