SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2201.00112 · ICCV · 2021

SurfGen: Adversarial 3D Shape Synthesis with Explicit Surface Discriminators

Tianqin Li, Andrew Luo, Tai Lee, Wen-Hao Zhang

arXiv · PDF · Open in the Atlas

Code that ran

5 of 8 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; 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/2201.00112.json · badge: /api/ran/2201.00112.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
aluo-x/neuralraycaster canonical (classifier label)
classifier basis not recorded on the link
5 ran · 3 no recorded run
FunctionStatusWhere it lives
get_centroids Ran
ran on a synthesized input
aluo-x/neuralraycaster/embree_rays.py
code served (permissive licence) · get_code("9862a9dbb2bf58dd")
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 numpy as np

torch.manual_seed(0)

# V should be a tensor of shape (num_vertices, vertex_dim)
# F should be a tensor of shape (num_faces, vertices_per_face) with indices into V
V = torch.randn(5, 4)  # 5 vertices, each with 4 dimensions
F = torch.tensor([
    [0, 1, 2],
    [1, 2, 3]
], dtype=torch.long)  # 2 faces, each with 3 vertex indices

ARGS = {
    "V": V,
    "F": F
}
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 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_cross_products Ran
ran on a synthesized input
aluo-x/neuralraycaster/embree_rays.py
code served (permissive licence) · get_code("54a8ffc4b17d8af5")
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

# Create vertex positions: 5 vertices, each with 3D coordinates
V = np.array([
    [0.0, 0.0, 0.0],
    [1.0, 0.0, 0.0],
    [0.0, 1.0, 0.0],
    [1.0, 1.0, 0.0],
    [0.5, 0.5, 1.0]
], dtype=np.float32)

# Create face indices: 2 triangular faces, each face has 3 vertex indices
F = np.array([
    [0, 1, 2],
    [1, 3, 4]
], dtype=np.int32)

ARGS = {
    "V": V,
    "F": F
}
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
make_list_parallel Ran
ran on a synthesized input
aluo-x/neuralraycaster/misc.py
code served (permissive licence) · get_code("1c5c57fb508ad27d")
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. 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 numpy as np
import torch
from torch_geometric.data import Data

torch.manual_seed(0)

# Create a single input_list dict with all required keys
input_list = {
    'depth_xyz': [
        np.random.randn(10, 3).astype(np.float32),  # radial depth xyz
        np.random.randn(8, 3).astype(np.float32),   # ortho depth xyz
    ],
    'depth_tri': [
        np.random.randint(0, 10, size=(5, 3)).astype(np.int64),  # radial depth tri
        np.random.randint(0, 8, size=(4, 3)).astype(np.int64),   # ortho depth tri
    ],
    'depth_ray': [
        np.random.randint(0, 100, size=(5,)).astype(np.int64),   # radial depth ray
        np.random.randint(0, 100, size=(4,)).astype(np.int64),   # ortho depth ray
    ],
    'prob_xyz_radial': [
        np.random.randn(6, 3).astype(np.float32),
        np.random.randn(7, 3).astype(np.float32),
    ],
    'prob_tri_radial': [
        np.random.randint(0, 10, size=(6, 3)).astype(np.int64),
        np.random.randint(0, 10, size=(7, 3)).astype(np.int64),
    ],
    'prob_ray_radial': [
        np.random.randint(0, 100, size=(6,)).astype(np.int64),
        np.random.randint(0, 100, size=(7,)).astype(np.int64),
    ],
    'prob_xyz_ortho': [
        np.random.randn(5, 3).astype(np.float32),
        np.random.randn(4, 3).astype(np.float32),
    ],
    'prob_tri_ortho': [
        np.random.randint(0, 10, size=(5, 3)).astype(np.int64),
        np.random.randint(0, 10, size=(4, 3)).astype(np.int64),
    ],
    'prob_ray_ortho': [
        np.random.randint(0, 100, size=(5,)).astype(np.int64),
        np.random.randint(0, 100, size=(4,)).astype(np.int64),
    ],
    'verts': np.random.randn(12, 3).astype(np.float32),
    'faces': np.random.randint(0, 12, size=(20, 3)).astype(np.int64),
}

# Create input_list_all as a list of input_lists
input_list_all = [input_list, input_list]

ARGS = {
    'input_list_all': input_list_all,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch, torch_geometric — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list 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
point_line_distance Ran
ran on a synthesized input
aluo-x/neuralraycaster/misc_differentiable.py
code served (permissive licence) · get_code("3c1be6b096c1eff2")
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

torch.manual_seed(0)

# p: a point tensor of shape (batch_size, dim)
# v0: start of line segment, shape (batch_size, dim)
# v1: end of line segment, shape (batch_size, dim)

batch_size = 2
dim = 4

p = torch.randn(batch_size, dim)
v0 = torch.randn(batch_size, dim)
v1 = torch.randn(batch_size, dim)

ARGS = {
    "p": p,
    "v0": v0,
    "v1": v1,
}
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 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
point_triangle_distance Ran
ran on a synthesized input
aluo-x/neuralraycaster/misc_differentiable.py
code served (permissive licence) · get_code("73eb14c99e282f04")
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

torch.manual_seed(0)

# Create small tensors for testing
# p: batch of points, shape (batch_size, dim)
# v0, v1, v2: batch of triangle vertices, shape (batch_size, dim)

batch_size = 2
dim = 4

p = torch.randn(batch_size, dim)
v0 = torch.randn(batch_size, dim)
v1 = torch.randn(batch_size, dim)
v2 = torch.randn(batch_size, dim)

ARGS = {
    "p": p,
    "v0": v0,
    "v1": v1,
    "v2": v2,
}
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 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
compute_bary_new Not yet run
unverified (harvested, no recorded run)
aluo-x/neuralraycaster/misc_differentiable.py
code served (permissive licence) · get_code("657f38f4a9a6737e")
How it ran

no invocation recorded.

expand_trig_np Not yet run
unverified (harvested, no recorded run)
aluo-x/neuralraycaster/misc.py
code served (permissive licence) · get_code("82f5c7cf8aa97a27")
How it ran

no invocation recorded.

get_face_areas Not yet run
unverified (harvested, no recorded run)
aluo-x/neuralraycaster/embree_rays.py
code served (permissive licence) · get_code("43f7bbe04a1a8cc0")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 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 8 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

Recent advances in deep generative models have led to immense progress in 3D shape synthesis. While existing models are able to synthesize shapes represented as voxels, point-clouds, or implicit functions, these methods only indirectly enforce the plausibility of the final 3D shape surface. Here we present a 3D shape synthesis framework (SurfGen) that directly applies adversarial training to the object surface. Our approach uses a differentiable spherical projection layer to capture and represent the explicit zero isosurface of an implicit 3D generator as functions defined on the unit sphere. By processing the spherical representation of 3D object surfaces with a spherical CNN in an adversarial setting, our generator can better learn the statistics of natural shape surfaces. We evaluate our model on large-scale shape datasets, and demonstrate that the end-to-end trained model is capable of generating high fidelity 3D shapes with diverse topology.

For agents

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

get_harvested_code_for_paper("2201.00112")
get_code_for_paper("2201.00112")
have("2201.00112")

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

curl https://syntology.ai/api/ran/2201.00112.json

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

[![Syntology run record](https://syntology.ai/api/ran/2201.00112.svg)](https://syntology.ai/paper/2201.00112)

Connect an agent — have() is free.