SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2003.13017 · 2020

Fast-MVSNet: Sparse-to-Dense Multi-View Stereo With Learned Propagation and Gauss-Newton Refinement

arXiv · PDF · Open in the Atlas

Code that ran

8 of 11 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; 0 honoured the contract we drafted, 0 violated it and 8 have no contract verdict; 0 failures of our instrument; 3 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/2003.13017.json · badge: /api/ran/2003.13017.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
svip-lab/FastMVSNet canonical (classifier label)
classifier basis: census
8 ran · 3 no recorded run
FunctionStatusWhere it lives
build_scheduler Ran
ran on a synthesized input
svip-lab/FastMVSNet/fastmvsnet/solver.py
code served (permissive licence) · get_code("cd22297339772e3e")
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 StepLR. 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
import torch.optim as optim

torch.manual_seed(0)

# Create a minimal model
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

model = MinimalModel()

# Create an optimizer
optimizer = optim.SGD(model.parameters(), lr=0.1)

# Create a config dict with the necessary structure
cfg = {
    'SCHEDULER': {
        'TYPE': 'StepLR',
        'StepLR': {
            'step_size': 10,
            'gamma': 0.1
        }
    }
}

# Convert to object-like access
class ConfigObj(dict):
    def __getattr__(self, name):
        if name in self:
            val = self[name]
            if isinstance(val, dict):
                return ConfigObj(val)
            return val
        raise AttributeError(f"No attribute {name}")
    
    def __setattr__(self, name, value):
        self[name] = value

cfg = ConfigObj(cfg)

ARGS = {
    'cfg': cfg,
    'optimizer': optimizer
}
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 StepLR — a StepLR output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a StepLR 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
cal_valid_less_percentage Ran
ran on a synthesized input
svip-lab/FastMVSNet/fastmvsnet/model.py
code served (permissive licence) · get_code("be79eaf9c24d8954")
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 for probing
batch_size = 2
height = 4
width = 3
num_images = 2

# pred_depth: predicted depth map
pred_depth = torch.rand(batch_size, 1, height, width) * 10.0

# gt_depth: ground truth depth map
gt_depth = torch.rand(batch_size, 1, height, width) * 10.0
# Add some zeros to create a valid mask
gt_depth[0, 0, 0, 0] = 0.0

# before_depth: depth from previous iteration (potentially different size)
before_depth = torch.rand(batch_size, 1, height // 2, width // 2) * 10.0

# depth_interval: interval values per image in batch
depth_interval = torch.tensor([0.5, 0.6])

# threshold: threshold for depth difference
threshold = 2.0

# valid_threshold: threshold for masking valid pixels
valid_threshold = 3.0

ARGS = {
    "pred_depth": pred_depth,
    "gt_depth": gt_depth,
    "before_depth": before_depth,
    "depth_interval": depth_interval,
    "threshold": threshold,
    "valid_threshold": valid_threshold,
}
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
encode_one_hot Ran
ran on a synthesized input
svip-lab/FastMVSNet/fastmvsnet/nn/functional.py
code served (permissive licence) · get_code("bb67d3c886157678")
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 torch.nn as nn

torch.manual_seed(0)

# Create a small target tensor with integer labels
# Shape: (N,) where N=2 (batch size)
target = torch.tensor([0, 2], dtype=torch.long)

# Number of classes
num_classes = 4

ARGS = {
    "target": target,
    "num_classes": num_classes,
}
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
get_pixel_grids Ran
ran on a synthesized input
svip-lab/FastMVSNet/fastmvsnet/functions/functions.py
code served (permissive licence) · get_code("c1d5bc61af47dd3d")
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, 12) 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)

ARGS = {
    "height": 4,
    "width": 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 2, dtype family float, shape (3, 12) — shape (3, 12) 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_propability_map Ran
ran on a synthesized input
svip-lab/FastMVSNet/fastmvsnet/functions/functions.py
code served (permissive licence) · get_code("d888261a28a10648")
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, 1, 3, 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)

batch_size = 2
height = 3
width = 4
num_depth_steps = 5
channels = 1

# cv: cost volume with shape (batch_size, depth_steps, height, width)
cv = torch.rand(batch_size, num_depth_steps, height, width)

# depth_map: estimated depth map with shape (batch_size, channels, height, width)
depth_map = torch.rand(batch_size, channels, height, width) * 2.0 + 1.0

# depth_start: starting depth value for each batch item
depth_start = torch.tensor([1.0, 1.5], dtype=torch.float32)

# depth_interval: depth interval for each batch item
depth_interval = torch.tensor([0.5, 0.5], dtype=torch.float32)

ARGS = {
    "cv": cv,
    "depth_map": depth_map,
    "depth_start": depth_start,
    "depth_interval": depth_interval,
}
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, 1, 3, 4) — shape (2, 1, 3, 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
group_weight Ran
ran on a synthesized input
svip-lab/FastMVSNet/fastmvsnet/solver.py
code served (permissive licence) · get_code("1e9b1e9a81221e61")
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)

# Build a small module with both regular parameters and batch norm parameters
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.bn = nn.BatchNorm1d(3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.bn(x)
        x = self.linear2(x)
        return x

# Instantiate the module
module = SimpleModel()

# Set weight_decay parameter
weight_decay = 1e-4

ARGS = {
    "module": module,
    "weight_decay": weight_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
pdist Ran
ran on a synthesized input
svip-lab/FastMVSNet/fastmvsnet/nn/functional.py
code served (permissive licence) · get_code("71629ec6951ead55")
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)

# Create a feature tensor with shape (batch_size, channels, num_features)
# batch_size=2, channels=4, num_features=3
feature = torch.randn(2, 4, 3)

ARGS = {
    "feature": feature
}
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
pdist2 Ran
ran on a synthesized input
svip-lab/FastMVSNet/fastmvsnet/nn/functional.py
code served (permissive licence) · get_code("af899c701bfaea6b")
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, 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)

# Create two feature tensors with shapes (batch_size, channels, num_features)
# batch_size=2, channels=4, num_features1=3, num_features2=5
feature1 = torch.randn(2, 4, 3)
feature2 = torch.randn(2, 4, 5)

ARGS = {
    "feature1": feature1,
    "feature2": feature2,
}
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, 5) — shape (2, 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
build_optimizer Not yet run
unverified (harvested, no recorded run)
svip-lab/FastMVSNet/fastmvsnet/solver.py
code served (permissive licence) · get_code("e70402c63e9d009c")
How it ran

no invocation recorded.

cal_less_percentage Not yet run
unverified (harvested, no recorded run)
svip-lab/FastMVSNet/fastmvsnet/model.py
code served (permissive licence) · get_code("4523bba800861a65")
How it ran

no invocation recorded.

load_cfg_from_file Not yet run
unverified (harvested, no recorded run)
svip-lab/FastMVSNet/fastmvsnet/config.py
code served (permissive licence) · get_code("1d7b97f7647a35a1")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 10 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
get_pixel_grids 1 other paper
1 later
Mi_Generalized_Binary_Search_Network_for_Highly-Efficient_Multi-View_Stereo_CVPR_2022_paper (2022 (year only))
Generalized Binary Search Network for Highly-Efficient Multi-View Stereo

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

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

Abstract

Almost all previous deep learning-based multi-view stereo (MVS) approaches focus on improving reconstruction quality. Besides quality, efficiency is also a desirable feature for MVS in real scenarios. Towards this end, this paper presents a Fast-MVSNet, a novel sparse-to-dense coarse-to-fine framework, for fast and accurate depth estimation in MVS. Specifically, in our Fast-MVSNet, we first construct a sparse cost volume for learning a sparse and high-resolution depth map. Then we leverage a small-scale convolutional neural network to encode the depth dependencies for pixels within a local region to densify the sparse high-resolution depth map. At last, a simple but efficient Gauss-Newton layer is proposed to further optimize the depth map. On one hand, the high-resolution depth map, the data-adaptive propagation method and the Gauss-Newton layer jointly guarantee the effectiveness of our method. On the other hand, all modules in our Fast-MVSNet are lightweight and thus guarantee the efficiency of our approach. Besides, our approach is also memory-friendly because of the sparse depth representation. Extensive experimental results show that our method is 5$\times$ and 14$\times$ faster than Point-MVSNet and R-MVSNet, respectively, while achieving comparable or even better results on the challenging Tanks and Temples dataset as well as the DTU dataset. Code is available at https://github.com/svip-lab/FastMVSNet.

For agents

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

get_harvested_code_for_paper("2003.13017")
get_code_for_paper("2003.13017")
have("2003.13017")

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

curl https://syntology.ai/api/ran/2003.13017.json

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

[![Syntology run record](https://syntology.ai/api/ran/2003.13017.svg)](https://syntology.ai/paper/2003.13017)

Connect an agent — have() is free.