SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 136970568 · ECCV · 2022

Learning an Isometric Surface Parameterization for Texture Unwrapping

Dimitris Samaras, Zhixin Shu, Ke Ma, Sagnik Das

Open in the Atlas

Code that ran

7 of 7 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, 1 violated it and 4 have no contract verdict; 2 failures of our instrument; 0 no recorded run), measured 2026-11-22; 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/136970568.json · badge: /api/ran/136970568.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
cvlab-stonybrook/Iso-UVField canonical (classifier label)
classifier basis: census
7 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_psnr Ran
ran on a synthesized input
cvlab-stonybrook/Iso-UVField/code/evaluation/unwarp_colmap.py
code served (permissive licence) · get_code("a1c624e9a4d54bf2")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import math

# Create small image arrays
np.random.seed(0)
img1 = np.random.randint(0, 256, size=(4, 3), dtype=np.uint8)
img2 = np.random.randint(0, 256, size=(4, 3), dtype=np.uint8)

# Create a mask with some True values
mask = np.random.rand(4, 3) > 0.5
mask = mask.astype(np.float64)

ARGS = {
    "img1": img1,
    "img2": img2,
    "mask": mask,
}
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 float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
compare_rotations Ran
ran on a synthesized input
cvlab-stonybrook/Iso-UVField/code/evaluation/unwarp_colmap.py
code served (permissive licence) · get_code("bfd88121e4dd2276")
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 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function expects two batches of rotation matrices
# R1 and R2 should be tensors of shape (batch_size, 3, 3)
# Each 3x3 matrix represents a rotation matrix

batch_size = 2
R1 = torch.randn(batch_size, 3, 3)
R2 = torch.randn(batch_size, 3, 3)

# Orthogonalize to make them closer to valid rotation matrices
# (not strictly necessary for the function to run, but more realistic)
U1, _, Vt1 = torch.linalg.svd(R1)
R1 = U1 @ Vt1

U2, _, Vt2 = torch.linalg.svd(R2)
R2 = U2 @ Vt2

ARGS = {
    "R1": R1,
    "R2": R2,
}
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 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_cameras_accuracy Ran
ran on a synthesized input
cvlab-stonybrook/Iso-UVField/code/evaluation/unwarp_colmap.py
code served (permissive licence) · get_code("1c5c2734a94354b3")
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
import numpy as np

torch.manual_seed(0)

# Create small batch of rotation matrices and translation vectors
# Batch size: 2, dimension: 3 (for 3D rotations)
n = 2
d = 3

# Create valid rotation matrices using QR decomposition
# pred_Rs: (2, 3, 3)
random_matrix_1 = torch.randn(n, d, d, dtype=torch.double)
pred_Rs, _ = torch.linalg.qr(random_matrix_1)
# Ensure det = 1 for proper rotation matrices
for i in range(n):
    if torch.det(pred_Rs[i]) < 0:
        pred_Rs[i, :, -1] *= -1

# gt_Rs: (2, 3, 3)
random_matrix_2 = torch.randn(n, d, d, dtype=torch.double)
gt_Rs, _ = torch.linalg.qr(random_matrix_2)
for i in range(n):
    if torch.det(gt_Rs[i]) < 0:
        gt_Rs[i, :, -1] *= -1

# pred_ts: (2, 3)
pred_ts = torch.randn(n, d, dtype=torch.double)

# gt_ts: (2, 3)
gt_ts = torch.randn(n, d, dtype=torch.double)

ARGS = {
    'pred_Rs': pred_Rs,
    'gt_Rs': gt_Rs,
    'pred_ts': pred_ts,
    'gt_ts': gt_ts,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cvxpy, numpy, 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
get_embedder Ran
ran on a synthesized input
cvlab-stonybrook/Iso-UVField/code/model/embedder.py
code served (permissive licence) · get_code("c08ac5a3b4184e50")
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 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

torch.manual_seed(0)

ARGS = {
    'multires': 4,
    'dim_in': 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 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
normalize Ran
ran, violated the contract we drafted
cvlab-stonybrook/Iso-UVField/code/datasets/doc_colmap_dataset.py
code served (permissive licence) · get_code("1d6f8390d799f055")
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 (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# The normalize function takes a single parameter 'x' which is expected to be
# a numpy array (or array-like). It divides x by its L2 norm.
# We need to provide a small numpy array as input.

x = np.array([3.0, 4.0])  # Simple 2D vector with norm=5, will normalize to [0.6, 0.8]

ARGS = {
    "x": x
}
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 (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
poses_avg Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cvlab-stonybrook/Iso-UVField/code/datasets/doc_colmap_dataset.py
code served (permissive licence) · get_code("d91b9b4ea99f0adc")
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, 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 numpy as np

# Analyze function usage:
# poses_avg takes 'poses' which is indexed as:
# - poses[0, :3, -1:] -> shape (3, 1)
# - poses[:, :3, 3] -> shape (N, 3)
# - poses[:, :3, 2] -> shape (N, 3)
# - poses[:, :3, 1] -> shape (N, 3)
# So poses needs shape (N, 3+1, 4) at minimum to support these indexing operations
# Let's use (2, 4, 4) to have distinct dimensions: batch=2, height=4, width=4

# Create a small poses array
poses = np.random.randn(2, 4, 4).astype(np.float32)

# Ensure the last column has valid height, width, focal data in first row
poses[0, :3, -1:] = np.array([[480], [640], [500]])

ARGS = {
    "poses": poses
}
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, 5) — shape (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
viewmatrix Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
cvlab-stonybrook/Iso-UVField/code/datasets/doc_colmap_dataset.py
code served (permissive licence) · get_code("aa52acee18f54d38")
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, 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

# Construct arguments for viewmatrix function
# The function expects:
# - z: a vector (used in normalize and cross product)
# - up: a vector (used in cross product)
# - pos: a vector (stacked into result)

# Create small 3D vectors as numpy arrays
z = np.array([0.0, 0.0, 1.0])
up = np.array([0.0, 1.0, 0.0])
pos = np.array([1.0, 2.0, 3.0])

ARGS = {
    "z": z,
    "up": up,
    "pos": pos
}
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, 4) — shape (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

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 3 later) and 4 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
normalize 19 other papers
4 dated earlier, 11 later, 4 same year, order not determinable
2003.08934 (2020-03)
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
2007.02442 (2020-07)
GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis
2104.04073 (2021-04)
Direct-PoseNet: Absolute Pose Regression with Photometric Consistency
and 16 more not named here
viewmatrix 17 other papers
4 dated earlier, 9 later, 4 same year, order not determinable
2003.08934 (2020-03)
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
2007.02442 (2020-07)
GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis
2104.04073 (2021-04)
Direct-PoseNet: Absolute Pose Regression with Photometric Consistency
and 14 more not named here
poses_avg 1 other paper
1 later
2401.04861 (2024-01)
CTNeRF: Cross-Time Transformer for Dynamic Neural Radiance Field from Monocular Video

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

In this paper, we present a novel approach to learn texture mapping for an isometrically deformed 3D surface and apply it for texture unwrapping of documents or other objects. Recent work on differentiable rendering techniques for implicit surfaces has shown highquality 3D scene reconstruction and view synthesis results. However, these methods typically learn the appearance color as a function of the surface points and lack explicit surface parameterization. Thus they do not allow texture map extraction or texture editing. We propose an efficient method to learn surface parameterization by learning a continuous bijective mapping between 3D surface positions and 2D texture-space coordinates. Our surface parameterization network can be conveniently plugged into a differentiable rendering pipeline and trained using multiview images and rendering loss. Using the learned parameterized implicit 3D surface we demonstrate state-of-the-art document-unwarping via texture extraction in both synthetic and real scenarios. We also show that our approach can reconstruct high-frequency textures for arbitrary objects. We further demonstrate the usefulness of our system by applying it to document and object texture editing. Code and related assets are available at: https://github.com/cvlab-stonybrook/Iso-UVField

For agents

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

get_harvested_code_for_paper("136970568")
get_code_for_paper("136970568")
have("136970568")

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

curl https://syntology.ai/api/ran/136970568.json

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

[![Syntology run record](https://syntology.ai/api/ran/136970568.svg)](https://syntology.ai/paper/136970568)

Connect an agent — have() is free.