SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1901.08971 · 2019

FaceForensics++: Learning to Detect Manipulated Facial Images

arXiv · PDF · Open in the Atlas

Code that ran

9 of 9 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; 4 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 0 failures of our instrument; 0 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/1901.08971.json · badge: /api/ran/1901.08971.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
ondyari/FaceForensics canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
twyunting/Deepfake_Video_Classifier community (classifier label)
classifier basis not recorded on the link
5 ran · 0 no recorded run
yyk-wew/F3Net reimplementation (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
FunctionStatusWhere it lives
DCT_mat Ran
ran, honoured the contract we drafted
yyk-wew/F3Net/models.py
pointer only (licence: NONE) · get_code("97170801bfeabdb7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yyk-wew/F3Net 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.

generate_filter Ran
ran, honoured the contract we drafted
yyk-wew/F3Net/models.py
pointer only (licence: NONE) · get_code("61e9294462032ca0")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yyk-wew/F3Net 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_all_image_files_read Ran
ran on a synthesized input
twyunting/Deepfake_Video_Classifier/code/HPC/HPC_construct_xfeatures_ylabels.py
code served (permissive licence) · get_code("d3ab7e011ae90f2f")
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 4, 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 cv2
import numpy as np
import tempfile
import os

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

# Create a few simple test images and save them
img1 = np.random.randint(0, 256, (10, 10, 3), dtype=np.uint8)
img2 = np.random.randint(0, 256, (10, 10, 3), dtype=np.uint8)

path1 = os.path.join(temp_dir, "image1.png")
path2 = os.path.join(temp_dir, "image2.png")

cv2.imwrite(path1, img1)
cv2.imwrite(path2, img2)

# Define arguments for the function
ARGS = {
    "path": temp_dir + "/",
    "filenames": ["image1.png", "image2.png"]
}
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 4, 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
get_boundingbox Ran
ran, honoured the contract we drafted
ondyari/FaceForensics/classification/detect_from_video.py
pointer only (licence: NOASSERTION) · get_code("cdd770694303a504")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, ondyari/FaceForensics (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.

get_list_filenames Ran
ran on a synthesized input
twyunting/Deepfake_Video_Classifier/code/HPC/HPC_construct_xfeatures_ylabels.py
code served (permissive licence) · get_code("06cdb128fdc7a477")
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 sys
import os
import glob
import tempfile
import shutil

# Create a temporary directory with some test files
temp_dir = tempfile.mkdtemp()
test_files = ['image1.jpg', 'image2.jpg', 'photo.jpg', 'document.txt']
for filename in test_files:
    filepath = os.path.join(temp_dir, filename)
    with open(filepath, 'w') as f:
        f.write('test content')

# Get the current working directory before any changes
cwd = os.getcwd()

ARGS = {
    'cwd': cwd,
    'path': temp_dir,
    'extension': 'jpg'
}
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
meanSubtraction Ran
ran on a synthesized input
twyunting/Deepfake_Video_Classifier/code/HPC/D_fake.py
code served (permissive licence) · get_code("faa064775cbc66a2")
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 (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 numpy as np

# The function takes an array/list of images and subtracts the mean from each
# Create a list of small numpy arrays to simulate image data
arr = [
    np.random.rand(4, 3).astype(np.float32),  # First image: 4x3
    np.random.rand(4, 3).astype(np.float32),  # Second image: 4x3
]

ARGS = {
    "arr": arr
}
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 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
meanSubtraction Ran
ran on a synthesized input
twyunting/Deepfake_Video_Classifier/code/HPC/D_real.py
code served (permissive licence) · get_code("8875205c75575c51")
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 (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 numpy as np

# Create a small batch of images as numpy arrays
# Using distinct dimensions: 2 images, each 4x3 (height x width)
batch_size = 2
img_height = 4
img_width = 3

# Create sample image data with different value ranges to show mean subtraction works
img1 = np.random.randint(0, 256, size=(img_height, img_width), dtype=np.uint8)
img2 = np.random.randint(0, 256, size=(img_height, img_width), dtype=np.uint8)

arr = [img1, img2]

ARGS = {
    "arr": arr
}
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 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
norm_sigma Ran
ran, honoured the contract we drafted
yyk-wew/F3Net/models.py
pointer only (licence: NONE) · get_code("22f1a0ed4d350586")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yyk-wew/F3Net 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.

svdTraining Ran
ran on a synthesized input
twyunting/Deepfake_Video_Classifier/code/HPC/D_fake.py
code served (permissive licence) · get_code("50149f89fd20f79e")
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

# The function expects arr to be indexable (arr[i]) for i in range(720)
# Each arr[i] should be a 2D array/matrix that can be passed to np.linalg.svd
# We'll create a list of 720 matrices with distinct shapes to make testing robust

arr = []
for i in range(720):
    # Create matrices with shape (4, 3) - small but rectangular for SVD interest
    matrix = np.random.randn(4, 3)
    arr.append(matrix)

ARGS = {
    "arr": arr
}
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

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 4 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 4 later) and 5 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
DCT_mat 3 other papers
3 later
2007.09355 (2020-07)
Thinking in Frequency: Face Forgery Detection by Mining Frequency-aware Clues
2104.09770 (2021-04)
M2TR: Multi-modal Multi-scale Transformers for Deepfake Detection
2405.19707 (2024-05)
DeMamba: AI-Generated Video Detection on Million-Scale GenVideo Benchmark
get_boundingbox 2 other papers
2 later
2307.07494 (2023-07)
TALL: Thumbnail Layout for Deepfake Video Detection
2403.02955 (2024-03)
XAI-Based Detection of Adversarial Attacks on Deepfake Detectors
norm_sigma 2 other papers
2 later
2007.09355 (2020-07)
Thinking in Frequency: Face Forgery Detection by Mining Frequency-aware Clues
2104.09770 (2021-04)
M2TR: Multi-modal Multi-scale Transformers for Deepfake Detection
generate_filter 1 other paper
1 later
2007.09355 (2020-07)
Thinking in Frequency: Face Forgery Detection by Mining Frequency-aware Clues

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 rapid progress in synthetic image generation and manipulation has now come to a point where it raises significant concerns for the implications towards society. At best, this leads to a loss of trust in digital content, but could potentially cause further harm by spreading false information or fake news. This paper examines the realism of state-of-the-art image manipulations, and how difficult it is to detect them, either automatically or by humans. To standardize the evaluation of detection methods, we propose an automated benchmark for facial manipulation detection. In particular, the benchmark is based on DeepFakes, Face2Face, FaceSwap and NeuralTextures as prominent representatives for facial manipulations at random compression level and size. The benchmark is publicly available and contains a hidden test set as well as a database of over 1.8 million manipulated images. This dataset is over an order of magnitude larger than comparable, publicly available, forgery datasets. Based on this data, we performed a thorough analysis of data-driven forgery detectors. We show that the use of additional domainspecific knowledge improves forgery detection to unprecedented accuracy, even in the presence of strong compression, and clearly outperforms human observers.

For agents

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

get_harvested_code_for_paper("1901.08971")
get_code_for_paper("1901.08971")
have("1901.08971")

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

curl https://syntology.ai/api/ran/1901.08971.json

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

[![Syntology run record](https://syntology.ai/api/ran/1901.08971.svg)](https://syntology.ai/paper/1901.08971)

Connect an agent — have() is free.