SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.11673 · 2024

MVSFormer++: Revealing the Devil in Transformer's Details for Multi-View Stereo

arXiv · PDF · Open in the Atlas

Code that ran

15 of 19 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, 1 violated it and 0 have no contract verdict; 11 failures of our instrument; 4 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2401.11673.json · badge: /api/ran/2401.11673.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
maybelx/mvsformerplusplus canonical (classifier label)
classifier basis: census
15 ran · 4 no recorded run
FunctionStatusWhere it lives
PositionEncoding3D Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
maybelx/mvsformerplusplus/models/position_encoding.py
code served (permissive licence) · get_code("9ccabd08bdaeae67")
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 5, dtype family float, shape (2, 24, 4, 3, 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
import math

torch.manual_seed(0)

# Construct position3d tensor with shape (B, 3, D, H, W)
# B=2 (batch), 3 coordinates (x, y, z), D=4, H=3, W=5
B = 2
D = 4
H = 3
W = 5
position3d = torch.randn(B, 3, D, H, W, dtype=torch.float32)

# C is the embedding dimension (must be even for sin/cos interleaving)
C = 8

# rescale is a float parameter
rescale = 4.0

ARGS = {
    "position3d": position3d,
    "C": C,
    "rescale": rescale,
}
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 5, dtype family float, shape (2, 24, 4, 3, 5) — shape (2, 24, 4, 3, 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
conf_regression Ran
ran; our contract draft was wrong, not the code
maybelx/mvsformerplusplus/models/module.py
code served (permissive licence) · get_code("d17a4985f968317d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 3, dtype family float, shape (2, 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 torch

torch.manual_seed(0)

# The function conf_regression takes two parameters: p and n
# p is used as a probability volume with shape (batch, ndepths, height, width)
# n is the window size for pooling, default is 4

# Create a small probability tensor
# Shape: (batch_size=2, ndepths=4, height=3, width=3)
# The values should be valid probabilities (sum to 1 along depth dimension)
p = torch.softmax(torch.randn(2, 4, 3, 3), dim=1)

# n is the pooling window size (default 4, but we'll use a smaller value for testing)
n = 4

ARGS = {
    "p": p,
    "n": n,
}
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 3, dtype family float, shape (2, 3, 3) — shape (2, 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
depth_regression Ran
ran, honoured the contract we drafted
maybelx/mvsformerplusplus/models/module.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from truongkhang/cds-mvsnet, is pointer only (licence: NONE) · get_code("16a0f9b77eb60de7")
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 truongkhang/cds-mvsnet, 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.

expand_as Ran
ran, violated the contract we drafted
maybelx/mvsformerplusplus/models/warping.py
code served (permissive licence) · get_code("79fd467dde213efb")
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, 1, 1) 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 tensors with different numbers of dimensions
# x has 2 dimensions, y has 4 dimensions
x = torch.randn(2, 4)  # shape: (2, 4)
y = torch.randn(2, 3, 5, 6)  # shape: (2, 3, 5, 6)

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, 4, 1, 1) — shape (2, 4, 1, 1) 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_layer_id_for_vit Ran
ran, honoured the contract we drafted
maybelx/mvsformerplusplus/models/lr_decay.py
code served (permissive licence) · get_code("d94135dc22062b8d")
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 int, ndim 0, dtype family int, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

# Construct valid arguments for get_layer_id_for_vit
# The function takes two parameters: name (str) and num_layers (int)

ARGS = {
    'name': 'blocks.5.attn.qkv',
    'num_layers': 12
}
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 int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_position_3d Ran
ran; our contract draft was wrong, not the code
maybelx/mvsformerplusplus/models/position_encoding.py
code served (permissive licence) · get_code("6dc3bfdaeb18a64f")
How it ran

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

import torch

torch.manual_seed(0)

# Parameters for get_position_3d
B = 2  # batch size
H = 4  # height
W = 3  # width
K_size = 3

# Intrinsic camera matrix (3x3)
K = torch.tensor([
    [500.0, 0.0, W/2],
    [0.0, 500.0, H/2],
    [0.0, 0.0, 1.0]
], dtype=torch.float32)

# Stack K for batch dimension
K = K.unsqueeze(0).repeat(B, 1, 1)

# Depth values: shape (B, num_depth)
num_depth = 5
depth_values = torch.linspace(0.5, 5.0, num_depth, dtype=torch.float32).unsqueeze(0).repeat(B, 1)

# Depth bounds
depth_min = 0.5
depth_max = 5.0

# Height, width bounds (None to auto-compute)
height_min = None
height_max = None
width_min = None
width_max = None

# Whether to normalize
normalize = True

ARGS = {
    'B': B,
    'H': H,
    'W': W,
    'K': K,
    'depth_values': depth_values,
    'depth_min': depth_min,
    'depth_max': depth_max,
    'height_min': height_min,
    'height_max': height_max,
    'width_min': width_min,
    'width_max': width_max,
    'normalize': normalize,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — 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
lift Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
maybelx/mvsformerplusplus/models/warping.py
code served (permissive licence) · get_code("2c4f405822853fe6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 3, dtype family float, shape (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 torch

torch.manual_seed(0)

# Build intrinsics matrix: shape (batch_size, 3, 3)
# This is a camera intrinsic matrix with fx, fy, cx, cy on the diagonal and offset
batch_size = 2
intrinsics = torch.tensor([
    [[500.0, 0.0, 320.0],
     [0.0, 500.0, 240.0],
     [0.0, 0.0, 1.0]],
    [[480.0, 0.0, 310.0],
     [0.0, 480.0, 235.0],
     [0.0, 0.0, 1.0]]
], dtype=torch.float32)

# x, y, z coordinates: shape (batch_size, num_points)
num_points = 4
x = torch.randn(batch_size, num_points)
y = torch.randn(batch_size, num_points)
z = torch.randn(batch_size, num_points)

# homogeneous flag
homogeneous = False

ARGS = {
    "x": x,
    "y": y,
    "z": z,
    "intrinsics": intrinsics,
    "homogeneous": homogeneous
}
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 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
make_nograd_func Ran
ran; our contract draft was wrong, not the code
maybelx/mvsformerplusplus/utils.py
code served (permissive licence) · get_code("a76fc7fb6958f97c")
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)

# Create a simple function that uses torch operations
def sample_func(x, y):
    # This function performs some torch operations
    result = x + y
    return result

# Test that the function works with gradients enabled
test_x = torch.randn(2, 4, requires_grad=True)
test_y = torch.randn(2, 4, requires_grad=True)

ARGS = {
    "func": sample_func
}
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
make_recursive_func Ran
ran; our contract draft was wrong, not the code
maybelx/mvsformerplusplus/utils.py
code served (permissive licence) · get_code("b62bec09ceeaca79")
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)

# Define a simple function to be made recursive
def example_func(x):
    """A simple function that doubles its input"""
    if isinstance(x, torch.Tensor):
        return x * 2
    elif isinstance(x, (int, float)):
        return x * 2
    else:
        return x

# The make_recursive_func takes a function as its only parameter
ARGS = {
    "func": example_func
}
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
param_groups_lrd Ran
ran; our contract draft was wrong, not the code
maybelx/mvsformerplusplus/models/lr_decay.py
code served (permissive licence) · get_code("904eb124a1c06ae8")
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 torch
import torch.nn as nn

torch.manual_seed(0)

class SimpleViT(nn.Module):
    def __init__(self, num_layers=3):
        super().__init__()
        self.cls_token = nn.Parameter(torch.randn(1, 1, 4))
        self.pos_embed = nn.Parameter(torch.randn(1, 5, 4))
        self.patch_embed = nn.Linear(3, 4)
        
        self.blocks = nn.ModuleList([
            nn.Sequential(
                nn.Linear(4, 4),
                nn.ReLU(),
                nn.Linear(4, 4)
            )
            for _ in range(num_layers)
        ])
        
        self.cross_blocks = nn.ModuleList([
            nn.Linear(4, 4)
            for _ in range(2)
        ])
        
        self.head = nn.Linear(4, 10)
        
        for name, param in self.named_parameters():
            param.requires_grad = True

model = SimpleViT(num_layers=3)

vit_lr = 0.001
weight_decay = 0.05
no_weight_decay_list = ['cls_token', 'pos_embed']
layer_decay = 0.75

ARGS = {
    'model': model,
    'vit_lr': vit_lr,
    'weight_decay': weight_decay,
    'no_weight_decay_list': no_weight_decay_list,
    'layer_decay': layer_decay
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family 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
parse_intrinsics Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
maybelx/mvsformerplusplus/models/warping.py
code served (permissive licence) · get_code("a06b72ce1e7e80f4")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function expects intrinsics to be a tensor that supports indexing
# with [batch_idx, row, col] to extract camera intrinsic matrix elements.
# It accesses: intrinsics[:, 0, 0], intrinsics[:, 1, 1], intrinsics[:, 0, 2], intrinsics[:, 1, 2]
# This is a batch of 3x3 camera intrinsic matrices.

batch_size = 2
intrinsics = torch.randn(batch_size, 3, 3)

ARGS = {
    "intrinsics": intrinsics
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
read_cameras_binary Ran
ran; our contract draft was wrong, not the code
maybelx/mvsformerplusplus/colmap2mvsnet.py
code served (permissive licence) · get_code("0bf4b207ce0d106a")
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 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 collections
import struct
import numpy as np
import tempfile
import os

CameraModel = collections.namedtuple('CameraModel', ['model_id', 'model_name', 'num_params'])

Camera = collections.namedtuple('Camera', ['id', 'model', 'width', 'height', 'params'])

CAMERA_MODELS = {CameraModel(model_id=0, model_name='SIMPLE_PINHOLE', num_params=3), CameraModel(model_id=1, model_name='PINHOLE', num_params=4), CameraModel(model_id=2, model_name='SIMPLE_RADIAL', num_params=4), CameraModel(model_id=3, model_name='RADIAL', num_params=5), CameraModel(model_id=4, model_name='OPENCV', num_params=8), CameraModel(model_id=5, model_name='OPENCV_FISHEYE', num_params=8), CameraModel(model_id=6, model_name='FULL_OPENCV', num_params=12), CameraModel(model_id=7, model_name='FOV', num_params=5), CameraModel(model_id=8, model_name='SIMPLE_RADIAL_FISHEYE', num_params=4), CameraModel(model_id=9, model_name='RADIAL_FISHEYE', num_params=5), CameraModel(model_id=10, model_name='THIN_PRISM_FISHEYE', num_params=12)}

CAMERA_MODEL_IDS = dict([(camera_model.model_id, camera_model) for camera_model in CAMERA_MODELS])

# Create a temporary binary file with valid camera data
temp_file = tempfile.NamedTemporaryFile(delete=False, suffix='.bin')
temp_path = temp_file.name

try:
    # Write number of cameras (2 cameras)
    num_cameras = 2
    temp_file.write(struct.pack('<Q', num_cameras))
    
    # Write first camera
    camera_id_1 = 1
    model_id_1 = 1  # PINHOLE with 4 params
    width_1 = 640
    height_1 = 480
    params_1 = [500.0, 500.0, 320.0, 240.0]  # fx, fy, cx, cy
    temp_file.write(struct.pack('<iiQQ', camera_id_1, model_id_1, width_1, height_1))
    temp_file.write(struct.pack('<' + 'd' * 4, *params_1))
    
    # Write second camera
    camera_id_2 = 2
    model_id_2 = 2  # SIMPLE_RADIAL with 4 params
    width_2 = 800
    height_2 = 600
    params_2 = [600.0, 400.0, 300.0, 0.1]  # fx, fy, cx, cy, k1
    temp_file.write(struct.pack('<iiQQ', camera_id_2, model_id_2, width_2, height_2))
    temp_file.write(struct.pack('<' + 'd' * 4, *params_2))
    
    temp_file.close()
    
    ARGS = {
        'path_to_model_file': temp_path
    }
    
finally:
    pass
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 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
read_cameras_text Ran
ran; our contract draft was wrong, not the code
maybelx/mvsformerplusplus/colmap2mvsnet.py
code served (permissive licence) · get_code("8a38e306ff4c5b42")
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 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 collections
import numpy as np
import tempfile
import os

Camera = collections.namedtuple('Camera', ['id', 'model', 'width', 'height', 'params'])

# Create a temporary file with valid camera data in the expected format
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_path = temp_file.name

# Write sample camera data
# Format: camera_id model width height params...
temp_file.write("# Camera list\n")
temp_file.write("1 SIMPLE_PINHOLE 640 480 320.5 240.5 500.0\n")
temp_file.write("2 PINHOLE 800 600 400.0 300.0 600.0 0.1\n")
temp_file.close()

ARGS = {
    'path': temp_path
}
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 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
read_next_bytes Ran
ran, honoured the contract we drafted
maybelx/mvsformerplusplus/colmap2mvsnet.py
code served (permissive licence) · get_code("56858e04e6fdb2ff")
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 tuple, 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 struct
import io

# Create a binary file-like object with test data
# We'll pack some values and then read them back
test_data = struct.pack('<hfi', 42, 3.14, 100)
fid = io.BytesIO(test_data)

ARGS = {
    'fid': fid,
    'num_bytes': 10,  # 2 bytes (h) + 4 bytes (f) + 4 bytes (i) = 10 bytes
    'format_char_sequence': 'hfi',
    'endian_character': '<'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
simple_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
maybelx/mvsformerplusplus/models/losses.py
code served (permissive licence) · get_code("312f59d39718ea4b")
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 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 torch.nn.functional as F

torch.manual_seed(0)

# Create small tensors with distinct dimensions
batch_size = 2
height = 4
width = 3

# outputs is a dict with 'depth' key
# depth_est should match shape of depth_gt and mask for masking operation
depth_est = torch.randn(batch_size, height, width)
outputs = {'depth': depth_est}

# depth_gt_ms: ground truth depth with same shape as depth_est
depth_gt_ms = torch.randn(batch_size, height, width)

# mask_ms: binary mask (will be compared with 0.5)
# Create a mask with some values > 0.5 and some <= 0.5
mask_ms = torch.rand(batch_size, height, width)

ARGS = {
    'outputs': outputs,
    'depth_gt_ms': depth_gt_ms,
    'mask_ms': mask_ms,
}
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 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
get_loss Not yet run
unverified (harvested, no recorded run)
maybelx/mvsformerplusplus/models/losses.py
code served (permissive licence) · get_code("aaa70aaf39de3e09")
How it ran

no invocation recorded.

get_multi_stage_losses Not yet run
unverified (harvested, no recorded run)
maybelx/mvsformerplusplus/models/losses.py
code served (permissive licence) · get_code("af0d20ff90759992")
How it ran

no invocation recorded.

init_range Not yet run
unverified (harvested, no recorded run)
maybelx/mvsformerplusplus/models/module.py
code served (permissive licence) · get_code("0499aba1289bdef7")
How it ran

no invocation recorded.

tensor2float Not yet run
unverified (harvested, no recorded run)
maybelx/mvsformerplusplus/utils.py
code served (permissive licence) · get_code("8e28f96374f1b285")
How it ran

no invocation recorded.

Also present in other papers' code

Of 19 samples lifted from this paper's linked repositories, 19 carry a function identity and 0 do not. Of the 19 distinct functions with an identity, 14 are also present in other papers whose code we hold (14 with at least one paper dated earlier, 3 the same month, 6 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
read_next_bytes 70 other papers
28 dated earlier, 2 the same month, 40 later
1804.00650 (2018-04)
DeepMVS: Learning Multi-view Stereopsis
1902.10556 (2019-02)
Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inference
1912.06378 (2019-12)
Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching
and 67 more not named here
read_cameras_text 37 other papers
21 dated earlier, 1 the same month, 15 later
1902.10556 (2019-02)
Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inference
1912.06378 (2019-12)
Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching
1912.11744 (2019-12)
Planar Prior Assisted PatchMatch Multi-View Stereo
and 34 more not named here
read_cameras_binary 21 other papers
13 dated earlier, 1 the same month, 7 later
1902.10556 (2019-02)
Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inference
1912.06378 (2019-12)
Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching
1912.11744 (2019-12)
Planar Prior Assisted PatchMatch Multi-View Stereo
and 18 more not named here
make_nograd_func 16 other papers
14 dated earlier, 2 later
1806.01260 (2018-06)
Digging Into Self-Supervised Monocular Depth Estimation
1912.06378 (2019-12)
Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching
2007.10872 (2020-07)
Dense Hybrid Recurrent Multi-view Stereo Net with Dynamic Consistency Checking
and 13 more not named here
make_recursive_func 14 other papers
13 dated earlier, 1 later
1912.06378 (2019-12)
Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching
2007.10872 (2020-07)
Dense Hybrid Recurrent Multi-view Stereo Net with Dynamic Consistency Checking
2104.00681 (2021-04)
NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Video
and 11 more not named here
tensor2float 10 other papers
9 dated earlier, 1 later
1912.06378 (2019-12)
Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching
2108.03824 (2021-08)
AA-RMVSNet: Adaptive Aggregation Recurrent Multi-view Stereo Network
2112.05126 (2021-12)
IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo
and 7 more not named here
depth_regression 5 other papers
5 dated earlier
1912.06378 (2019-12)
Cascade Cost Volume for High-Resolution Multi-View Stereo and Stereo Matching
2112.05999 (2021-12)
Curvature-guided dynamic scale networks for Multi-view Stereo
2201.01501 (2022-01)
Rethinking Depth Estimation for Multi-View Stereo: A Unified Representation
and 2 more not named here
conf_regression 1 other paper
1 dated earlier
2208.02541 (2022-08)
MVSFormer: Multi-View Stereo by Learning Robust Image Features and Temperature-based Depth
expand_as 1 other paper
1 dated earlier
2208.02541 (2022-08)
MVSFormer: Multi-View Stereo by Learning Robust Image Features and Temperature-based Depth
get_layer_id_for_vit 1 other paper
1 dated earlier
2208.02541 (2022-08)
MVSFormer: Multi-View Stereo by Learning Robust Image Features and Temperature-based Depth
lift 1 other paper
1 dated earlier
2208.02541 (2022-08)
MVSFormer: Multi-View Stereo by Learning Robust Image Features and Temperature-based Depth
param_groups_lrd 1 other paper
1 dated earlier
2208.02541 (2022-08)
MVSFormer: Multi-View Stereo by Learning Robust Image Features and Temperature-based Depth
parse_intrinsics 1 other paper
1 dated earlier
2208.02541 (2022-08)
MVSFormer: Multi-View Stereo by Learning Robust Image Features and Temperature-based Depth
simple_loss 1 other paper
1 dated earlier
2208.02541 (2022-08)
MVSFormer: Multi-View Stereo by Learning Robust Image Features and Temperature-based Depth

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

Recent advancements in learning-based Multi-View Stereo (MVS) methods have prominently featured transformer-based models with attention mechanisms. However, existing approaches have not thoroughly investigated the profound influence of transformers on different MVS modules, resulting in limited depth estimation capabilities. In this paper, we introduce MVSFormer++, a method that prudently maximizes the inherent characteristics of attention to enhance various components of the MVS pipeline. Formally, our approach involves infusing cross-view information into the pre-trained DINOv2 model to facilitate MVS learning. Furthermore, we employ different attention mechanisms for the feature encoder and cost volume regularization, focusing on feature and spatial aggregations respectively. Additionally, we uncover that some design details would substantially impact the performance of transformer modules in MVS, including normalized 3D positional encoding, adaptive attention scaling, and the position of layer normalization. Comprehensive experiments on DTU, Tanks-and-Temples, BlendedMVS, and ETH3D validate the effectiveness of the proposed method. Notably, MVSFormer++ achieves state-of-the-art performance on the challenging DTU and Tanks-and-Temples benchmarks.

For agents

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

get_harvested_code_for_paper("2401.11673")
get_code_for_paper("2401.11673")
have("2401.11673")

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

curl https://syntology.ai/api/ran/2401.11673.json

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

[![Syntology run record](https://syntology.ai/api/ran/2401.11673.svg)](https://syntology.ai/paper/2401.11673)

Connect an agent — have() is free.