SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.03407 · 2024

Bilateral Reference for High-Resolution Dichotomous Image Segmentation

arXiv · PDF · Open in the Atlas

Code that ran

13 of 14 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; 3 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 10 failures of our instrument; 1 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/2401.03407.json · badge: /api/ran/2401.03407.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
zhengpeng7/birefnet canonical (classifier label)
classifier basis: census
13 ran · 1 no recorded run
FunctionStatusWhere it lives
FB_blur_fusion_foreground_estimator_cpu_2 Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zhengpeng7/birefnet/image_proc.py
code served (permissive licence) · get_code("e7965221ba2e5453")
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
import cv2
from PIL import Image

# Create small test data
height, width = 4, 3
image = np.random.rand(height, width, 3).astype(np.float32)
alpha = np.random.rand(height, width).astype(np.float32)

# Ensure alpha values are in [0, 1]
alpha = np.clip(alpha, 0, 1)

ARGS = {
    'image': image,
    'alpha': alpha,
    'r': 90
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, 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 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
SSIM Ran
ran; our contract draft was wrong, not the code
zhengpeng7/birefnet/loss.py
code served (permissive licence) · get_code("32a9ebc9a8eebc49")
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, 5) 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 torch import nn

torch.manual_seed(0)

# Create small sample tensors for x and y
# SSIM expects image tensors, typically (batch, channels, height, width)
# Using batch=2, channels=3, height=4, width=5 for distinct dimensions
x = torch.randn(2, 3, 4, 5)
y = torch.randn(2, 3, 4, 5)

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 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 4, 5) 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
check_state_dict Ran
ran; our contract draft was wrong, not the code
zhengpeng7/birefnet/utils.py
code served (permissive licence) · get_code("a0f2af79dd5536ac")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# The function takes a state_dict (a dict mapping strings to tensors/values)
# and optionally unwanted_prefixes (a list of strings).
# It removes unwanted prefixes from the keys in the state_dict.

# Create a sample state_dict with various prefixes
state_dict = {
    'module.layer1.weight': torch.randn(4, 3),
    'module.layer1.bias': torch.randn(4),
    '_orig_mod.layer2.weight': torch.randn(3, 2),
    'layer3.weight': torch.randn(2, 4),
    'module._orig_mod.layer4.bias': torch.randn(5),
}

unwanted_prefixes = ['module.', '_orig_mod.']

ARGS = {
    'state_dict': state_dict,
    'unwanted_prefixes': unwanted_prefixes,
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
create_window Ran
ran, honoured the contract we drafted
zhengpeng7/birefnet/loss.py
code served (permissive licence) · get_code("6154e3744ece5728")
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 (3, 1, 5, 5) 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 torch.autograd import Variable
from math import exp

torch.manual_seed(0)

def gaussian(window_size, sigma):
    gauss = torch.Tensor([exp(-(x - window_size // 2) ** 2 / float(2 * sigma ** 2)) for x in range(window_size)])
    return gauss / gauss.sum()

def create_window(window_size, channel):
    _1D_window = gaussian(window_size, 1.5).unsqueeze(1)
    _2D_window = _1D_window.mm(_1D_window.t()).float().unsqueeze(0).unsqueeze(0)
    window = Variable(_2D_window.expand(channel, 1, window_size, window_size).contiguous())
    return window

ARGS = {
    "window_size": 5,
    "channel": 3
}
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 (3, 1, 5, 5) — shape (3, 1, 5, 5) 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 Ran
ran, honoured the contract we drafted
zhengpeng7/birefnet/loss.py
code served (permissive licence) · get_code("c56b7ef16f309a45")
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 Tensor, ndim 1, dtype family float, shape (5,) 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 math import exp

torch.manual_seed(0)

ARGS = {
    "window_size": 5,
    "sigma": 1.0,
}
Setup block drafted by
a language model, Claude Haiku 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 1, dtype family float, shape (5,) — shape (5,) 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
generate_smoothed_gt Ran
ran, honoured the contract we drafted
zhengpeng7/birefnet/utils.py
code served (permissive licence) · get_code("625f65061a0463a9")
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 ground truth tensor
# The function expects a tensor and performs element-wise operations on it
gts = torch.randn(2, 4)

ARGS = {
    "gts": gts
}
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
image2patches Ran
ran; our contract draft was wrong, not the code
zhengpeng7/birefnet/models/birefnet.py
code served (permissive licence) · get_code("939c02df4fca5af1")
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 (8, 4, 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
from einops import rearrange

torch.manual_seed(0)

# Create a small image tensor: batch=2, channels=4, height=8, width=6
image = torch.randn(2, 4, 8, 6)

# Create a patch_ref tensor with shape that divides image dimensions
# image is (2, 4, 8, 6), so patch_ref could be (2, 4, 4, 3) to get grid_h=2, grid_w=2
patch_ref = torch.randn(2, 4, 4, 3)

# Define the transformation string
transformation = 'b c (hg h) (wg w) -> (b hg wg) c h w'

ARGS = {
    'image': image,
    'grid_h': 2,
    'grid_w': 2,
    'patch_ref': patch_ref,
    'transformation': transformation
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, 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 (8, 4, 4, 3) — shape (8, 4, 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
load_images_parallel Ran
ran; our contract draft was wrong, not the code
zhengpeng7/birefnet/evaluation/metrics.py
code served (permissive licence) · get_code("a769e9d2c7633ce9")
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
import numpy as np
import cv2

# Create temporary directory with test images
temp_dir = tempfile.mkdtemp()

# Create a simple grayscale ground truth image
gt_img = np.random.randint(0, 256, (100, 100), dtype=np.uint8)
gt_path_1 = os.path.join(temp_dir, 'gt_1.png')
cv2.imwrite(gt_path_1, gt_img)

gt_img_2 = np.random.randint(0, 256, (120, 150), dtype=np.uint8)
gt_path_2 = os.path.join(temp_dir, 'gt_2.png')
cv2.imwrite(gt_path_2, gt_img_2)

# Create corresponding prediction images
pred_img = np.random.randint(0, 256, (100, 100), dtype=np.uint8)
pred_path_1 = os.path.join(temp_dir, 'pred_1.png')
cv2.imwrite(pred_path_1, pred_img)

pred_img_2 = np.random.randint(0, 256, (150, 200), dtype=np.uint8)
pred_path_2 = os.path.join(temp_dir, 'pred_2.png')
cv2.imwrite(pred_path_2, pred_img_2)

gt_paths = [gt_path_1, gt_path_2]
pred_paths = [pred_path_1, pred_path_2]

ARGS = {
    'gt_paths': gt_paths,
    'pred_paths': pred_paths,
    'num_workers': 2,
    'verbose': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy, tqdm — 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
load_single_image_pair Ran
ran; our contract draft was wrong, not the code
zhengpeng7/birefnet/evaluation/metrics.py
code served (permissive licence) · get_code("fb4f4bbd0eb2e5b6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import cv2
import numpy as np
import tempfile

# Create temporary directory and test image files
tmpdir = tempfile.mkdtemp()

# Create a small grayscale ground truth image
gt_image = np.random.randint(0, 256, (100, 80), dtype=np.uint8)
gt_path = os.path.join(tmpdir, 'gt_image.png')
cv2.imwrite(gt_path, gt_image)

# Create a small grayscale prediction image
pred_image = np.random.randint(0, 256, (120, 100), dtype=np.uint8)
pred_path = os.path.join(tmpdir, 'pred_image.jpg')
cv2.imwrite(pred_path, pred_image)

# Prepare arguments
args = (gt_path, pred_path, 0)

ARGS = {
    'args': args
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
mean_blur Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zhengpeng7/birefnet/image_proc.py
code served (permissive licence) · get_code("76639f9e42708b4e")
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, 5) 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)

# Construct a small input tensor [B, C, H, W]
# Using distinct sizes: batch=2, channels=3, height=4, width=5
x = torch.randn(2, 3, 4, 5)

# kernel_size should be a positive integer
kernel_size = 3

ARGS = {
    "x": x,
    "kernel_size": kernel_size,
}
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, 4, 5) — shape (2, 3, 4, 5) 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
patches2image Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zhengpeng7/birefnet/models/birefnet.py
code served (permissive licence) · get_code("05e78beaf2a1a37a")
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, 6, 6) 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
from einops import rearrange

torch.manual_seed(0)

# Create patches tensor
# Shape: (b*hg*wg, c, h, w) where b=2, hg=2, wg=2, c=4, h=3, w=3
patches = torch.randn(8, 4, 3, 3)

# Parameters
grid_h = 2
grid_w = 2
patch_ref = None
transformation = '(b hg wg) c h w -> b c (hg h) (wg w)'

ARGS = {
    'patches': patches,
    'grid_h': grid_h,
    'grid_w': grid_w,
    'patch_ref': patch_ref,
    'transformation': transformation
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, 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, 4, 6, 6) — shape (2, 4, 6, 6) 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
path_to_image Ran
ran; our contract draft was wrong, not the code
zhengpeng7/birefnet/utils.py
code served (permissive licence) · get_code("f66e6359cff25940")
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 Image. 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 cv2
import numpy as np
from PIL import Image
import tempfile
import os

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

# Create a simple RGB image and save it
test_image_array = np.random.randint(0, 256, (512, 512, 3), dtype=np.uint8)
test_image_bgr = cv2.cvtColor(test_image_array, cv2.COLOR_RGB2BGR)
cv2.imwrite(temp_image_path, test_image_bgr)

ARGS = {
    'path': temp_image_path,
    'size': (1024, 1024),
    'color_type': 'rgb'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
process_metrics_single Ran
ran; our contract draft was wrong, not the code
zhengpeng7/birefnet/evaluation/metrics.py
code served (permissive licence) · get_code("dcd0a21071a6862c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
from unittest.mock import Mock

# Create mock measure objects that have a step method
class MockMeasure:
    def __init__(self):
        pass
    
    def step(self, **kwargs):
        pass

# Create data dictionary with required keys
data = {
    'gt': np.random.randint(0, 256, size=(32, 32), dtype=np.uint8),
    'pred': np.random.randint(0, 256, size=(32, 32), dtype=np.uint8),
    'gt_path': '/tmp/gt/sample.png'
}

# Create measures dictionary with mock metric objects
measures = {
    'E': MockMeasure(),
    'S': MockMeasure(),
    'F': MockMeasure(),
    'MAE': MockMeasure(),
    'MSE': MockMeasure(),
    'WF': MockMeasure(),
    'HCE': MockMeasure(),
    'MBA': MockMeasure(),
    'BIoU': MockMeasure()
}

# Create metrics list with a subset of available metrics
metrics = ['E', 'S', 'F', 'MAE', 'MSE', 'WF', 'HCE', 'MBA', 'BIoU']

ARGS = {
    'data': data,
    'measures': measures,
    'metrics': metrics
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
FB_blur_fusion_foreground_estimator_cpu Not yet run
unverified (harvested, no recorded run)
zhengpeng7/birefnet/image_proc.py
code served (permissive licence) · get_code("0799073430991c59")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 2 the same month, 2 later) and 11 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
gaussian 160 other papers
64 dated earlier, 4 the same month, 91 later, 1 same year, order not determinable · 7 only as an assembly brick
1812.11941 (2018-12)
High Quality Monocular Depth Estimation via Transfer Learning
1904.01538 (2019-04)
Spatial Attentive Single-Image Deraining with a High Quality Real Rain Dataset
1907.11628 (2019-07)
Unsupervised Learning for Optical Flow Estimation Using Pyramid Convolution LSTM
and 157 more not named here
create_window 52 other papers
27 dated earlier, 2 the same month, 23 later · 5 only as an assembly brick
1904.01538 (2019-04)
Spatial Attentive Single-Image Deraining with a High Quality Real Rain Dataset
1907.11628 (2019-07)
Unsupervised Learning for Optical Flow Estimation Using Pyramid Convolution LSTM
2002.11616 (2020-02)
Zooming Slow-Mo: Fast and Accurate One-Stage Space-Time Video Super-Resolution
and 49 more not named here
generate_smoothed_gt 1 other paper
1 dated earlier
2302.14485 (2023-02)
Memory-Aided Contrastive Consensus Learning for Co-salient Object Detection

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

We introduce a novel bilateral reference framework (BiRefNet) for high-resolution dichotomous image segmentation (DIS). It comprises two essential components: the localization module (LM) and the reconstruction module (RM) with our proposed bilateral reference (BiRef). The LM aids in object localization using global semantic information. Within the RM, we utilize BiRef for the reconstruction process, where hierarchical patches of images provide the source reference and gradient maps serve as the target reference. These components collaborate to generate the final predicted maps. We also introduce auxiliary gradient supervision to enhance focus on regions with finer details. Furthermore, we outline practical training strategies tailored for DIS to improve map quality and training process. To validate the general applicability of our approach, we conduct extensive experiments on four tasks to evince that BiRefNet exhibits remarkable performance, outperforming task-specific cutting-edge methods across all benchmarks. Our codes are available at https://github.com/ZhengPeng7/BiRefNet.

For agents

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

get_harvested_code_for_paper("2401.03407")
get_code_for_paper("2401.03407")
have("2401.03407")

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

curl https://syntology.ai/api/ran/2401.03407.json

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

[![Syntology run record](https://syntology.ai/api/ran/2401.03407.svg)](https://syntology.ai/paper/2401.03407)

Connect an agent — have() is free.