SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2003.06814 · 2020

Towards Face Encryption by Generating Adversarial Identity Masks

arXiv · PDF · Open in the Atlas

Code that ran

15 of 23 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; 2 honoured the contract we drafted, 0 violated it and 9 have no contract verdict; 4 failures of our instrument; 8 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/2003.06814.json · badge: /api/ran/2003.06814.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
shawnxyang/tip-im canonical (classifier label)
classifier basis: census
15 ran · 8 no recorded run
FunctionStatusWhere it lives
IOU Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
shawnxyang/tip-im/align_methods/align.py
code served (permissive licence) · get_code("9348919b3a8be935")
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 float, 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
import numpy as np

torch.manual_seed(0)

# The function expects two bounding boxes in format [x1, y1, x2, y2]
# where (x1, y1) is top-left corner and (x2, y2) is bottom-right corner

# Create two overlapping bounding boxes
Reframe = [10.0, 20.0, 50.0, 60.0]  # Box 1: top-left (10, 20), bottom-right (50, 60)
GTframe = [30.0, 40.0, 70.0, 80.0]  # Box 2: top-left (30, 40), bottom-right (70, 80)

ARGS = {
    'Reframe': Reframe,
    'GTframe': GTframe
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, 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
LResNet50E_IR Ran
ran on a synthesized input
shawnxyang/tip-im/networks/CosFace.py
code served (permissive licence) · get_code("32918e149a737ef6")
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 LResNet. 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 torch.nn as nn

torch.manual_seed(0)

# LResNet50E_IR takes only one parameter: is_gray
# We'll test both False (RGB) and True (grayscale) cases
# Let's use is_gray=False as the default

ARGS = {
    'is_gray': False
}
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 LResNet — a LResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a LResNet 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
affine Ran
ran on a synthesized input
shawnxyang/tip-im/input_diversify.py
code served (permissive licence) · get_code("00e18ccd3f1f30c4")
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, 4, 2, 2) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# x: input tensor - grid_sample expects (N, C, H, W) format
# Create a small batch with 2 samples, 4 channels, 3x3 spatial dims
x = torch.randn(2, 4, 3, 3)

# vgrid: sampling grid for grid_sample
# grid_sample expects (N, H_out, W_out, 2) format with normalized coordinates in [-1, 1]
# Create a small grid for 2x2 output spatial dimensions
vgrid = torch.rand(2, 2, 2, 2) * 2 - 1  # normalized to [-1, 1]

# device: string specifying device - use 'cpu' for CPU-only environment
device = 'cpu'

ARGS = {
    'x': x,
    'vgrid': vgrid,
    'device': device
}
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, 4, 2, 2) — shape (2, 4, 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
gaussian_kernel_matrix Ran
ran on a synthesized input
shawnxyang/tip-im/mmd.py
code served (permissive licence) · get_code("00f2288642d682fa")
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 (3, 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 torch

torch.manual_seed(0)

# x: matrix of shape (n_samples, n_features)
x = torch.randn(2, 4)

# y: matrix of shape (m_samples, n_features) - must have same feature dim as x
y = torch.randn(3, 4)

# sigmas: 1D tensor of sigma values for gaussian kernel
sigmas = torch.tensor([1.0, 2.0, 0.5])

ARGS = {
    'x': x,
    'y': y,
    '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 (3, 2) — shape (3, 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
get_block Ran
ran, honoured the contract we drafted
shawnxyang/tip-im/networks/ArcFace.py
code served (permissive licence) · get_code("ad5dbf57f3ea2633")
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 list, ndim 2, 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
from collections import namedtuple

torch.manual_seed(0)

# The function get_block takes in_channel, depth, num_units, and stride
# It returns a list of Bottleneck namedtuples

ARGS = {
    'in_channel': 64,
    'depth': 128,
    'num_units': 3,
    'stride': 2
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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 2, 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_blocks Ran
ran; our contract draft was wrong, not the code
shawnxyang/tip-im/networks/ArcFace.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from VladAndronik/face_swap, is pointer only (licence: NONE) · get_code("ea7941eb1ea8ffb2")
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 VladAndronik/face_swap, 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_dataset_celeb Ran
ran on a synthesized input
shawnxyang/tip-im/align_methods/face_image.py
code served (permissive licence) · get_code("924034eebb04536f")
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 os
import tempfile
from easydict import EasyDict as edict

# Create a temporary directory and a clean_list.txt file
temp_dir = tempfile.mkdtemp()
input_dir = temp_dir + '/celeb_data'
os.makedirs(input_dir, exist_ok=True)

# Create the clean_list.txt file with sample data
clean_list_file = input_dir + '_clean_list.txt'
with open(clean_list_file, 'w') as f:
    f.write('./m.person1/image1.jpg\n')
    f.write('./m.person1/image2.jpg\n')
    f.write('./m.person2/image3.jpg\n')
    f.write('invalid_line.jpg\n')
    f.write('./m.person2/image4.jpg\n')

# Create the actual image files (as dummy files)
os.makedirs(os.path.join(input_dir, 'm.person1'), exist_ok=True)
os.makedirs(os.path.join(input_dir, 'm.person2'), exist_ok=True)
open(os.path.join(input_dir, 'm.person1/image1.jpg'), 'w').close()
open(os.path.join(input_dir, 'm.person1/image2.jpg'), 'w').close()
open(os.path.join(input_dir, 'm.person2/image3.jpg'), 'w').close()
open(os.path.join(input_dir, 'm.person2/image4.jpg'), 'w').close()

ARGS = {
    'input_dir': input_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
easydict — 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
get_dataset_webface Ran
ran on a synthesized input
shawnxyang/tip-im/align_methods/face_image.py
code served (permissive licence) · get_code("baa6487e16d2b4a8")
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
from easydict import EasyDict as edict

# Create a temporary directory and clean_list.txt file for testing
temp_dir = tempfile.mkdtemp()
input_dir = temp_dir + "/webface"
os.makedirs(input_dir, exist_ok=True)

# Create the clean_list.txt file with sample data
clean_list_file = input_dir + '_clean_list.txt'
with open(clean_list_file, 'w') as f:
    f.write("person1/image1.jpg 0\n")
    f.write("person2/image2.jpg 1\n")
    f.write("person1/image3.jpg 0\n")

# Create dummy image files so the paths are valid
os.makedirs(os.path.join(input_dir, "person1"), exist_ok=True)
os.makedirs(os.path.join(input_dir, "person2"), exist_ok=True)
open(os.path.join(input_dir, "person1/image1.jpg"), 'w').close()
open(os.path.join(input_dir, "person2/image2.jpg"), 'w').close()
open(os.path.join(input_dir, "person1/image3.jpg"), 'w').close()

ARGS = {
    'input_dir': input_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
easydict — 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
l2_norm Ran
ran; our contract draft was wrong, not the code
shawnxyang/tip-im/networks/ArcFace.py
code served (permissive licence) · get_code("c54fea429589425d")
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 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)

input_tensor = torch.randn(2, 4)
axis = 1

ARGS = {
    'input': input_tensor,
    'axis': axis
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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
layer Ran
ran; our contract draft was wrong, not the code
shawnxyang/tip-im/align_methods/detect_face.py
code served (permissive licence) · get_code("fab32b621c712604")
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 function. 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 torch.nn as nn

torch.manual_seed(0)

class MockNetworkBuilder:
    """Mock class that simulates a network builder with layer management."""
    def __init__(self):
        self.terminals = []
        self.layers = {}
        self._name_counter = {}
    
    def get_unique_name(self, name):
        """Generate unique layer names."""
        if name not in self._name_counter:
            self._name_counter[name] = 0
        else:
            self._name_counter[name] += 1
        return f"{name}_{self._name_counter[name]}"
    
    def feed(self, layer_output):
        """Update terminals with new layer output."""
        self.terminals = [layer_output]

class MockOp:
    """Mock operation function that processes layer input."""
    def __call__(self, builder, layer_input, name=None):
        # Simulate a simple operation that returns a tensor
        if isinstance(layer_input, list):
            return torch.cat(layer_input, dim=1)
        else:
            return layer_input * 2

# Create a mock builder instance
mock_builder = MockNetworkBuilder()

# Initialize terminals with a sample tensor (batch_size=2, features=4)
sample_tensor = torch.randn(2, 4)
mock_builder.terminals = [sample_tensor]

# Create a mock operation
mock_op = MockOp()

# Define ARGS
ARGS = {
    'op': mock_op
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
load_property Ran
ran on a synthesized input
shawnxyang/tip-im/align_methods/face_image.py
code served (permissive licence) · get_code("a0e97123727a9fb0")
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 EasyDict. 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 os
import tempfile
from easydict import EasyDict as edict

# Create a temporary directory with a 'property' file
temp_dir = tempfile.mkdtemp()
property_file = os.path.join(temp_dir, 'property')

# Write a valid property file with num_classes and image dimensions
with open(property_file, 'w') as f:
    f.write('10,224,224\n')

ARGS = {
    'data_dir': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
easydict — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EasyDict — a EasyDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a EasyDict 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
myphi Ran
ran on a synthesized input
shawnxyang/tip-im/networks/SphereFace.py
code served (permissive licence) · get_code("4edba53421463c1a")
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 math

torch.manual_seed(0)

ARGS = {
    "x": torch.tensor([[0.1, 0.2, 0.3, 0.4], [0.5, 0.6, 0.7, 0.8]]),
    "m": 2
}
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
pairwise_distance Ran
ran on a synthesized input
shawnxyang/tip-im/mmd.py
code served (permissive licence) · get_code("21454b651fb3987f")
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 (3, 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 torch

torch.manual_seed(0)

# Create two 2D tensors with the same number of features (dim 1)
# x: shape (2, 4) - 2 samples, 4 features
# y: shape (3, 4) - 3 samples, 4 features
x = torch.randn(2, 4)
y = torch.randn(3, 4)

ARGS = {
    'x': x,
    'y': y
}
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 (3, 2) — shape (3, 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
preprocess Ran
ran on a synthesized input
shawnxyang/tip-im/align_methods/face_preprocess.py
code served (permissive licence) · get_code("1586a591b369e039")
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 cv2

# Create a minimal test image (3 channels, height=200, width=160)
test_img = np.random.randint(0, 256, (200, 160, 3), dtype=np.uint8)

# Create landmarks for face alignment (5 facial landmarks)
test_landmarks = np.array([
    [30.2946, 51.6963],
    [65.5318, 51.5014],
    [48.0252, 71.7366],
    [33.5493, 92.3655],
    [62.7299, 92.2041]
], dtype=np.float32)

ARGS = {
    'img': test_img,
    'image_size': (112, 112),
    'bbox': None,
    'landmark': test_landmarks
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy, skimage — 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
to_rgb Ran
ran, honoured the contract we drafted
shawnxyang/tip-im/align_methods/align.py
code served (permissive licence) · get_code("29507ad0ee182d8c")
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 ndarray, ndim 3, dtype family int, shape (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 numpy as np

# Create a small grayscale image (2D array)
img = np.array([[10, 20, 30, 40],
                [50, 60, 70, 80],
                [90, 100, 110, 120]], dtype=np.uint8)

ARGS = {
    'img': img
}
Setup block drafted by
a language model, Claude Sonnet 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 3, dtype family int, shape (3, 4, 3) — shape (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
RandomCrop Not yet run
unverified (harvested, no recorded run)
shawnxyang/tip-im/input_diversify.py
code served (permissive licence) · get_code("86226494dfd08ce2")
How it ran

no invocation recorded.

Resize Not yet run
unverified (harvested, no recorded run)
shawnxyang/tip-im/input_diversify.py
code served (permissive licence) · get_code("919b0ccbc0ed8a63")
How it ran

no invocation recorded.

create_mtcnn Not yet run
unverified (harvested, no recorded run)
shawnxyang/tip-im/align_methods/detect_face.py
code served (permissive licence) · get_code("edfa6042244eb012")
How it ran

no invocation recorded.

detect_face Not yet run
unverified (harvested, no recorded run)
shawnxyang/tip-im/align_methods/detect_face.py
code served (permissive licence) · get_code("96bf1c60f74fc61d")
How it ran

no invocation recorded.

get_model Not yet run
unverified (harvested, no recorded run)
shawnxyang/tip-im/networks/FaceModel.py
code served (permissive licence) · get_code("01506ca0db5cb1e7")
How it ran

no invocation recorded.

maximum_mean_discrepancy Not yet run
unverified (harvested, no recorded run)
shawnxyang/tip-im/mmd.py
code served (permissive licence) · get_code("d848d80fdf341607")
How it ran

no invocation recorded.

parse_lst_line Not yet run
unverified (harvested, no recorded run)
shawnxyang/tip-im/align_methods/face_preprocess.py
code served (permissive licence) · get_code("82b10a215c8cd30f")
How it ran

no invocation recorded.

read_image Not yet run
unverified (harvested, no recorded run)
shawnxyang/tip-im/align_methods/face_preprocess.py
code served (permissive licence) · get_code("79625c2830b046be")
How it ran

no invocation recorded.

Also present in other papers' code

Of 23 samples lifted from this paper's linked repositories, 23 carry a function identity and 0 do not. Of the 23 distinct functions with an identity, 16 are also present in other papers whose code we hold (15 with at least one paper dated earlier, 2 the same month, 12 later) and 7 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
l2_norm 29 other papers
5 dated earlier, 24 later · 2 only as an assembly brick
1804.07573 (2018-04)
MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
1905.11946 (2019-05)
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
and 26 more not named here
get_block 27 other papers
4 dated earlier, 1 the same month, 22 later
1804.07573 (2018-04)
MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
and 24 more not named here
get_blocks 16 other papers
4 dated earlier, 1 the same month, 11 later
1804.07573 (2018-04)
MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
and 13 more not named here
layer 5 other papers
4 dated earlier, 1 later
1606.00373 (2016-06)
Deeper Depth Prediction with Fully Convolutional Residual Networks
1611.08050 (2016-11)
Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
1808.00157 (2018-08)
Instance-level Human Parsing via Part Grouping Network
and 2 more not named here
IOU 3 other papers
3 dated earlier
1801.07698 (2018-01)
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
1905.00641 (2019-05)
RetinaFace: Single-stage Dense Face Localisation in the Wild
to_rgb 3 other papers
3 dated earlier
1503.03832 (2015-03)
FaceNet: A Unified Embedding for Face Recognition and Clustering
1801.07698 (2018-01)
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
create_mtcnn 2 other papers
1 dated earlier, 1 later
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
2007.04118 (2020-07)
RobFR: Benchmarking Adversarial Robustness on Face Recognition
detect_face 2 other papers
1 dated earlier, 1 later
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
2007.04118 (2020-07)
RobFR: Benchmarking Adversarial Robustness on Face Recognition
get_dataset_celeb 2 other papers
1 dated earlier, 1 later
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
2007.04118 (2020-07)
RobFR: Benchmarking Adversarial Robustness on Face Recognition
get_dataset_webface 2 other papers
1 dated earlier, 1 later
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
2007.04118 (2020-07)
RobFR: Benchmarking Adversarial Robustness on Face Recognition
load_property 2 other papers
1 dated earlier, 1 later
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
2007.04118 (2020-07)
RobFR: Benchmarking Adversarial Robustness on Face Recognition
myphi 2 other papers
2 dated earlier
1411.7923 (2014-11)
Learning Face Representation from Scratch
2002.12186 (2020-02)
University-1652: A Multi-view Multi-source Benchmark for Drone-based Geo-localization
parse_lst_line 2 other papers
1 dated earlier, 1 later
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
2007.04118 (2020-07)
RobFR: Benchmarking Adversarial Robustness on Face Recognition
read_image 2 other papers
1 dated earlier, 1 later
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
2007.04118 (2020-07)
RobFR: Benchmarking Adversarial Robustness on Face Recognition
LResNet50E_IR 1 other paper
1 dated earlier
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
preprocess 1 other paper
1 later
2007.04118 (2020-07)
RobFR: Benchmarking Adversarial Robustness on Face Recognition

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

As billions of personal data being shared through social media and network, the data privacy and security have drawn an increasing attention. Several attempts have been made to alleviate the leakage of identity information from face photos, with the aid of, e.g., image obfuscation techniques. However, most of the present results are either perceptually unsatisfactory or ineffective against face recognition systems. Our goal in this paper is to develop a technique that can encrypt the personal photos such that they can protect users from unauthorized face recognition systems but remain visually identical to the original version for human beings. To achieve this, we propose a targeted identity-protection iterative method (TIP-IM) to generate adversarial identity masks which can be overlaid on facial images, such that the original identities can be concealed without sacrificing the visual quality. Extensive experiments demonstrate that TIP-IM provides 95\%+ protection success rate against various state-of-the-art face recognition models under practical test scenarios. Besides, we also show the practical and effective applicability of our method on a commercial API service.

For agents

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

get_harvested_code_for_paper("2003.06814")
get_code_for_paper("2003.06814")
have("2003.06814")

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

curl https://syntology.ai/api/ran/2003.06814.json

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

[![Syntology run record](https://syntology.ai/api/ran/2003.06814.svg)](https://syntology.ai/paper/2003.06814)

Connect an agent — have() is free.