SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1907.12849 · 2019

Orientation-aware Semantic Segmentation on Icosahedron Spheres

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 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 5 have no contract verdict; 0 failures of our instrument; 0 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/1907.12849.json · badge: /api/ran/1907.12849.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
matsuren/HexRUNet_pytorch community (classifier label)
classifier basis not recorded on the link
5 ran · 0 no recorded run
FunctionStatusWhere it lives
genuv Ran
ran on a synthesized input
matsuren/HexRUNet_pytorch/utils/projection_helper.py
code served (permissive licence) · get_code("ce42e7036f068561")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 3, dtype family float, shape (8, 16, 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 function genuv takes three parameters: h, w, and v_rot
# h: height (int)
# w: width (int)
# v_rot: vertical rotation (float, between -pi/2 and pi/2)

# Based on the function body:
# - h and w are used in np.arange and division operations
# - v_rot is used in comparisons and trigonometric operations
# - The function is cached via lru_cache, so arguments must be hashable
# - h and w should be integers (used with np.arange)
# - v_rot should be a float

ARGS = {
    "h": 8,
    "w": 16,
    "v_rot": 0.0
}
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 3, dtype family float, shape (8, 16, 2) — shape (8, 16, 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_unfold_icosahedron Ran
ran on a synthesized input
matsuren/HexRUNet_pytorch/utils/geometry_helper.py
code served (permissive licence) · get_code("f11924d473e9d2bd")
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 numpy as np

ARGS = {
    "level": 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 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_unfold_imgcoord Ran
ran on a synthesized input
matsuren/HexRUNet_pytorch/utils/geometry_helper.py
code served (permissive licence) · get_code("1b3db7ec8d750bd8")
How it ran

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

import numpy as np
from copy import copy

def get_base_icosahedron():
    t = (1.0 + 5.0 ** 0.5) / 2.0
    vertices = [-1, t, 0, 1, t, 0, 0, 1, t, -t, 0, 1, -t, 0, -1, 0, 1, -t, t, 0, -1, t, 0, 1, 0, -1, t, -1, -t, 0, 0, -1, -t, 1, -t, 0]
    faces = [0, 2, 1, 0, 3, 2, 0, 4, 3, 0, 5, 4, 0, 1, 5, 1, 7, 6, 1, 2, 7, 2, 8, 7, 2, 3, 8, 3, 9, 8, 3, 4, 9, 4, 10, 9, 4, 5, 10, 5, 6, 10, 5, 1, 6, 6, 7, 11, 7, 8, 11, 8, 9, 11, 9, 10, 11, 10, 6, 11]
    vertices = np.reshape(vertices, (-1, 3)) / (np.sin(2 * np.pi / 5) * 2)
    faces = np.reshape(faces, (-1, 3))
    ry = -vertices[0]
    rx = np.cross(ry, vertices[1])
    rx /= np.linalg.norm(rx)
    rz = np.cross(rx, ry)
    R = np.stack([rx, ry, rz])
    vertices = vertices.dot(R.T)
    return (vertices, faces)

def get_base_unfold():
    v, f = get_base_icosahedron()
    unfold_v = {i: [] for i in range(12)}
    edge_len = 1 / np.sin(2 * np.pi / 5)
    h = 3 ** 0.5 * edge_len / 2
    for i in range(5):
        unfold_v[0].append([i * edge_len, 0])
    for _ in range(5):
        unfold_v[1].append([-0.5 * edge_len, h])
    unfold_v[1][1] = [-0.5 * edge_len + 5 * edge_len, h]
    unfold_v[1][4] = [-0.5 * edge_len + 5 * edge_len, h]
    for i in range(2, 6):
        for _ in range(5):
            unfold_v[i].append([(0.5 + i - 2) * edge_len, h])
    for _ in range(5):
        unfold_v[6].append([-edge_len, 2 * h])
    unfold_v[6][1] = [-edge_len + 5 * edge_len, 2 * h]
    unfold_v[6][2] = [-edge_len + 5 * edge_len, 2 * h]
    unfold_v[6][4] = [-edge_len + 5 * edge_len, 2 * h]
    for i in range(7, 11):
        for _ in range(5):
            unfold_v[i].append([(i - 7) * edge_len, 2 * h])
    for i in range(5):
        unfold_v[11].append([(-0.5 + i) * edge_len, 3 * h])
    for i in range(len(unfold_v)):
        unfold_v[i] = np.array(unfold_v[i])
    return (unfold_v, f)

unfold_v, f = get_base_unfold()

ARGS = {
    "level": 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 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
uv2xyz Ran
ran on a synthesized input
matsuren/HexRUNet_pytorch/utils/projection_helper.py
code served (permissive licence) · get_code("a7bc3ed9660ca10a")
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, 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 uv2xyz takes a uv array and performs trigonometric operations
# It expects uv to have shape (..., 2) where the last dimension has 2 elements
# representing u and v angles

# Create a simple uv array with shape (2, 2) - batch of 2 UV coordinates
uv = np.array([
    [0.5, 1.0],      # First UV pair
    [1.5, 2.0]       # Second UV pair
], dtype=np.float32)

ARGS = {
    "uv": uv
}
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, 3) — shape (2, 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
xyz2uv Ran
ran on a synthesized input
matsuren/HexRUNet_pytorch/utils/projection_helper.py
code served (permissive licence) · get_code("354420e492429761")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 3, dtype family float, shape (2, 4, 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 function xyz2uv takes xyz coordinates and converts them to spherical coordinates (u, v)
# xyz should be an array-like with shape (..., 3) where the last dimension contains x, y, z

# Create a small test input with shape (2, 4, 3)
# This represents 2 batches of 4 points, each with (x, y, z) coordinates
xyz_input = np.random.randn(2, 4, 3).astype(np.float32)

ARGS = {
    "xyz": xyz_input
}
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 3, dtype family float, shape (2, 4, 2) — shape (2, 4, 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

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 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.

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

Repositories linked to this paper

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

Abstract

We address semantic segmentation on omnidirectional images, to leverage a holistic understanding of the surrounding scene for applications like autonomous driving systems. For the spherical domain, several methods recently adopt an icosahedron mesh, but systems are typically rotation invariant or require significant memory and parameters, thus enabling execution only at very low resolutions. In our work, we propose an orientation-aware CNN framework for the icosahedron mesh. Our representation allows for fast network operations, as our design simplifies to standard network operations of classical CNNs, but under consideration of north-aligned kernel convolutions for features on the sphere. We implement our representation and demonstrate its memory efficiency up-to a level-8 resolution mesh (equivalent to 640 x 1024 equirectangular images). Finally, since our kernels operate on the tangent of the sphere, standard feature weights, pretrained on perspective data, can be directly transferred with only small need for weight refinement. In our evaluation our orientation-aware CNN becomes a new state of the art for the recent 2D3DS dataset, and our Omni-SYNTHIA version of SYNTHIA. Rotation invariant classification and segmentation tasks are additionally presented for comparison to prior art.

For agents

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

get_harvested_code_for_paper("1907.12849")
get_code_for_paper("1907.12849")
have("1907.12849")

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

curl https://syntology.ai/api/ran/1907.12849.json

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

[![Syntology run record](https://syntology.ai/api/ran/1907.12849.svg)](https://syntology.ai/paper/1907.12849)

Connect an agent — have() is free.