SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.18842 · CVPR · 2024

ViewFusion: Towards Multi-View Consistency via Interpolated Denoising

Anton Van Den Hengel, Xianghui Yang, Sameera Ramasinghe, Yan Zuo, Gil Avraham, Loris Bazzani

arXiv · PDF · Open in the Atlas

Code that ran

14 of 17 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; 6 honoured the contract we drafted, 2 violated it and 0 have no contract verdict; 6 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/2402.18842.json · badge: /api/ran/2402.18842.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
Wi-sc/ViewFusion canonical (classifier label)
classifier basis: census
14 ran · 3 no recorded run
FunctionStatusWhere it lives
add_image_to_canvas Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Wi-sc/ViewFusion/vis_gen_rendering.py
code served (permissive licence) · get_code("6cd164825eab4981")
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 4, dtype family int, shape (100, 512, 1024, 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

# Create a minimal canvas for testing
model_name_list = ['B07DB8XGY2', 'B07P5LNMDK']
n_rows = len(model_name_list)
n_columns = 4
box = 256

# Initialize canvas with the expected shape
canvas = np.ones((100, n_rows * box, n_columns * box, 3), dtype=np.uint8) * 255

# Create a test image (224x224 RGB image)
test_img = np.random.randint(0, 256, (224, 224, 3), dtype=np.uint8)

# Define arguments for add_image_to_canvas
ARGS = {
    'frame_id': 0,
    'img': test_img,
    'canvas': canvas,
    'row_id': 0,
    'col_id': 0
}
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 ndarray, ndim 4, dtype family int, shape (100, 512, 1024, 3) — shape (100, 512, 1024, 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
get_K Ran
ran, honoured the contract we drafted
Wi-sc/ViewFusion/3drec/pose.py
code served (permissive licence) · get_code("870d0b3c55043f70")
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 2, dtype family float, shape (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

# The function get_K takes three parameters:
# - H: height (integer)
# - W: width (integer)  
# - FoV_x: field of view in x direction (float, in degrees)

# These are used in simple arithmetic operations and numpy functions
# No complex objects needed

ARGS = {
    "H": 480,
    "W": 640,
    "FoV_x": 45.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (3, 3) — shape (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
gradient_x Ran
ran, violated the contract we drafted
Wi-sc/ViewFusion/3drec/my3d.py
code served (permissive licence) · get_code("1be56d4101a6952b")
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, 2, 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)

# The function takes an image tensor and computes gradient in x direction
# by subtracting adjacent columns. It expects a 4D tensor (batch, channels, height, width)
# or a 2D tensor (batch, width) based on the slicing operation [:, :-1] and [:, 1:]

# Create a small image tensor: batch_size=2, channels=3, height=4, width=5
img = torch.randn(2, 3, 4, 5)

ARGS = {
    "img": img
}
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, 2, 4, 5) — shape (2, 2, 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
gradient_y Ran
ran, honoured the contract we drafted
Wi-sc/ViewFusion/3drec/my3d.py
code served (permissive licence) · get_code("71912f3e0d290417")
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, 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 small 2D image tensor
# The function computes img[:-1, :] - img[1:, :]
# So we need a tensor with at least 2 rows
img = torch.randn(4, 3)

ARGS = {
    "img": img
}
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, 3) — shape (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
imgs_to_torch Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Wi-sc/ViewFusion/3drec/misc.py
code served (permissive licence) · get_code("bd3c8ca13d7ff5e2")
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, 64, 64) 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 torch

torch.manual_seed(0)

# Create a small batch of images
# Shape: (N, H, W, C) = (2, 64, 64, 3)
# dtype must be uint8
imgs = np.random.randint(0, 256, size=(2, 64, 64, 3), dtype=np.uint8)

ARGS = {
    'imgs': imgs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
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, 3, 64, 64) — shape (2, 3, 64, 64) 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
karras_t_schedule Ran
ran, honoured the contract we drafted
Wi-sc/ViewFusion/3drec/adapt.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from pals-ttic/sjc, is pointer only (licence: NOASSERTION) · get_code("8285678722ad9b94")
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 pals-ttic/sjc, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

load_model_from_config Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Wi-sc/ViewFusion/3drec/adapt_sd.py
code served (permissive licence) · get_code("e30fc050b1d17697")
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 StandInModel. 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 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)

# Stand-in config class with a model attribute
class StandInConfig:
    def __init__(self):
        self.model = StandInModelConfig()

class StandInModelConfig:
    """Minimal config that can be instantiated"""
    pass

# Minimal model class
class StandInModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)
    
    def load_state_dict(self, state_dict, strict=False):
        """Mock load_state_dict that returns missing and unexpected keys"""
        missing_keys = []
        unexpected_keys = []
        return missing_keys, unexpected_keys

# Mock instantiate_from_config function
def mock_instantiate_from_config(config):
    return StandInModel()

# Patch the ldm.util module
import sys
from types import ModuleType

ldm = ModuleType('ldm')
ldm_util = ModuleType('ldm.util')
ldm_util.instantiate_from_config = mock_instantiate_from_config
ldm.util = ldm_util
sys.modules['ldm'] = ldm
sys.modules['ldm.util'] = ldm_util

# Create a minimal checkpoint file content
config = StandInConfig()

# Create checkpoint data with required structure
checkpoint_data = {
    'global_step': 42,
    'state_dict': {
        'linear.weight': torch.randn(3, 4),
        'linear.bias': torch.randn(3)
    }
}

# We need to mock torch.load to return our checkpoint without file I/O
original_torch_load = torch.load

def mock_torch_load(path, map_location=None):
    return checkpoint_data

torch.load = mock_torch_load

ARGS = {
    'config': config,
    'ckpt': 'dummy_checkpoint.pt',
    'verbose': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
ldm, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type StandInModel — a StandInModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a StandInModel 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_np_imgs Ran
ran, honoured the contract we drafted
Wi-sc/ViewFusion/3drec/run_img_sampling.py
code served (permissive licence) · get_code("b6b6f9d6e69e2075")
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 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 tempfile
import numpy as np
from pathlib import Path

# Create a temporary directory and files for testing
temp_dir = tempfile.mkdtemp()

# Create a .npy file
npy_array = np.random.rand(2, 4).astype(np.float32)
npy_path = Path(temp_dir) / "test_image.npy"
np.save(npy_path, npy_array)

# Create a .npz file
npz_path = Path(temp_dir) / "test_image.npz"
np.savez(npz_path, npy_array)

# Define ARGS with both test cases
ARGS = {
    "fname": str(npy_path)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 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
normalize Ran
ran, violated the contract we drafted
Wi-sc/ViewFusion/3drec/my3d.py
code served (permissive licence) · get_code("498fc1ddd4b3b1ee")
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 1, dtype family float, shape (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 numpy as np

torch_seed = 0

# The function 'normalize' takes a single parameter 'v' (a vector)
# and returns v divided by its norm (L2 norm by default)
# v should be a numpy array or array-like

v = np.array([1.0, 2.0, 3.0, 4.0])

ARGS = {
    "v": v
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (4,) — shape (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
pick_out_cfgs Ran
ran; our contract draft was wrong, not the code
Wi-sc/ViewFusion/3drec/adapt_gddpm.py
code served (permissive licence) · get_code("87a0ac3d40fc695c")
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 numpy as np

torch.manual_seed(0)

# The function pick_out_cfgs takes a source dict and a list of keys,
# and returns a new dict with only those keys from the source.
# It uses: indexing src[k] for each k in target_ks

src = {
    'learning_rate': 0.001,
    'batch_size': 32,
    'epochs': 100,
    'optimizer': 'adam',
    'hidden_dim': 256,
    'dropout': 0.1,
    'weight_decay': 1e-5
}

target_ks = ['learning_rate', 'batch_size', 'epochs', 'optimizer']

ARGS = {
    'src': src,
    'target_ks': target_ks
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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
power_schedule Ran
ran, honoured the contract we drafted
Wi-sc/ViewFusion/3drec/adapt.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from pals-ttic/sjc, is pointer only (licence: NOASSERTION) · get_code("c5ab2a66337ec866")
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 pals-ttic/sjc, and GitHub could not classify the licence of that repository (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

sample_near_eye Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Wi-sc/ViewFusion/3drec/pose.py
code served (permissive licence) · get_code("376462d7f357bca4")
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 1, dtype family float, shape (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

# Set random seed for reproducibility
np.random.seed(0)

# The function sample_near_eye takes:
# - eye: a numpy array representing an eye position (used in addition with sample_random_vector output)
# - norm: a float parameter controlling the magnitude of the random perturbation (default 0.05)

# Create a small eye position vector (3D coordinates)
eye = np.array([1.0, 2.0, 3.0])

ARGS = {
    "eye": eye,
    "norm": 0.05
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (3,) — shape (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
sample_random_vector Ran
ran, honoured the contract we drafted
Wi-sc/ViewFusion/3drec/pose.py
code served (permissive licence) · get_code("cc66a0e5b806f055")
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 1, dtype family float, shape (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

def sample_random_vector(norm=0.05):
    vec = np.random.randn(3)
    vec /= np.linalg.norm(vec)
    vec *= norm
    return vec

ARGS = {
    "norm": 0.05
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (3,) — shape (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
torch_samps_to_imgs Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Wi-sc/ViewFusion/3drec/misc.py
code served (permissive licence) · get_code("590cd2734f888f2a")
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 4, dtype family int, shape (2, 4, 5, 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 tiny batch of images in the expected format
# Shape: (batch_size=2, channels=3, height=4, width=5)
# Values in range [-1, 1] as expected by the function
imgs = torch.randn(2, 3, 4, 5)

ARGS = {
    "imgs": imgs,
    "uncenter": True,
}
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 ndarray, ndim 4, dtype family int, shape (2, 4, 5, 3) — shape (2, 4, 5, 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_ckpt Not yet run
unverified (harvested, no recorded run)
Wi-sc/ViewFusion/3drec/adapt_gddpm.py
code served (permissive licence) · get_code("16f5658b4f119a9d")
How it ran

no invocation recorded.

load_objaverse_model Not yet run
unverified (harvested, no recorded run)
Wi-sc/ViewFusion/3drec/adapt_sd.py
code served (permissive licence) · get_code("21aee206d05b291f")
How it ran

no invocation recorded.

restore_checkpoint Not yet run
unverified (harvested, no recorded run)
Wi-sc/ViewFusion/3drec/adapt_vesde.py
code served (permissive licence) · get_code("4c2866ac682fd5c2")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 6 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 0 the same month, 0 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
karras_t_schedule 3 other papers
3 dated earlier
2206.00364 (2022-06)
Elucidating the Design Space of Diffusion-Based Generative Models
2212.00774 (2022-12)
Score Jacobian Chaining: Lifting Pretrained 2D Diffusion Models for 3D Generation
2304.10261 (2023-04)
Anything-3D: Towards Single-view Anything Reconstruction in the Wild
power_schedule 3 other papers
3 dated earlier
2206.00364 (2022-06)
Elucidating the Design Space of Diffusion-Based Generative Models
2212.00774 (2022-12)
Score Jacobian Chaining: Lifting Pretrained 2D Diffusion Models for 3D Generation
2304.10261 (2023-04)
Anything-3D: Towards Single-view Anything Reconstruction in the Wild
get_K 1 other paper
1 dated earlier
2304.10261 (2023-04)
Anything-3D: Towards Single-view Anything Reconstruction in the Wild
imgs_to_torch 1 other paper
1 dated earlier
2304.10261 (2023-04)
Anything-3D: Towards Single-view Anything Reconstruction in the Wild
normalize 1 other paper
1 dated earlier
2304.10261 (2023-04)
Anything-3D: Towards Single-view Anything Reconstruction in the Wild
torch_samps_to_imgs 1 other paper
1 dated earlier
2304.10261 (2023-04)
Anything-3D: Towards Single-view Anything Reconstruction in the Wild

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

Novel-view synthesis through diffusion models has demonstrated remarkable potential for generating diverse and high-quality images. Yet, the independent process of image generation in these prevailing methods leads to challenges in maintaining multiple view consistency. To address this, we introduce ViewFusion, a novel, trainingfree algorithm that can be seamlessly integrated into existing pre-trained diffusion models. Our approach adopts an auto-regressive method that implicitly leverages previously generated views as context for next view generation, ensuring robust multi-view consistency during the novelview generation process. Through a diffusion process that fuses known-view information via interpolated denoising, our framework successfully extends single-view conditioned models to work in multiple-view conditional settings without any additional fine-tuning. Extensive experimental results demonstrate the effectiveness of ViewFusion in generating consistent and detailed novel views.

For agents

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

get_harvested_code_for_paper("2402.18842")
get_code_for_paper("2402.18842")
have("2402.18842")

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

curl https://syntology.ai/api/ran/2402.18842.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.18842.svg)](https://syntology.ai/paper/2402.18842)

Connect an agent — have() is free.