SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2111.00969 · 2021

Generative Occupancy Fields for 3D Surface-Aware Image Synthesis

arXiv · PDF · Open in the Atlas

Code that ran

7 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 6 have no contract verdict; 1 failure of our instrument; 4 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/2111.00969.json · badge: /api/ran/2111.00969.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
sheldontsui/gof_neurips2021 canonical (classifier label)
classifier basis not recorded on the link
7 ran · 4 no recorded run
FunctionStatusWhere it lives
create_samples Ran
ran; our contract draft was wrong, not the code
sheldontsui/gof_neurips2021/extract_shapes.py
code served (permissive licence) · get_code("306b4b17e95e50e5")
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)

ARGS = {
    "N": 4,
    "voxel_origin": [0, 0, 0],
    "cube_length": 2.0,
}
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 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
ellipse Ran
ran on a synthesized input
sheldontsui/gof_neurips2021/render_meshvideo.py
code served (permissive licence) · get_code("3859bc14d69a3ada")
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 math

ARGS = {
    "i": 0,
    "num": 10,
    "x_max": 5.0,
    "y_max": 3.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
ellipse Ran
ran on a synthesized input
sheldontsui/gof_neurips2021/render_video.py
code served (permissive licence) · get_code("e21ac6d18e16ad07")
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 math

ARGS = {
    'i': 3,
    'num': 10,
    'x_max': 2.5,
    'y_max': 1.8,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
get_current_step Ran
ran on a synthesized input
sheldontsui/gof_neurips2021/curriculums.py
code served (permissive licence) · get_code("a551faf28fc86a9f")
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 int, ndim 0, dtype family int, 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

torch.manual_seed(0)

ARGS = {
    'curriculum': {
        'update_epochs': [0, 5, 10, 15]
    },
    'epoch': 7
}
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 int, ndim 0, dtype family int, 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
img_normalize Ran
ran on a synthesized input
sheldontsui/gof_neurips2021/render_video.py
code served (permissive licence) · get_code("360b46dceb9d86ad")
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, 4, 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)

# img_normalize takes an image tensor and normalizes it
# It clamps values to [-1, 1], then rescales to [0, 1]
# The function expects a tensor as input

# Create a small test image tensor with values in range [-1, 1]
img = torch.randn(2, 4, 3, 3)  # batch=2, channels=4, height=3, width=3
img = torch.clamp(img, min=-1, max=1)  # ensure it's in valid range

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, 4, 3, 3) — shape (2, 4, 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
last_upsample_step Ran
ran on a synthesized input
sheldontsui/gof_neurips2021/curriculums.py
code served (permissive licence) · get_code("31312e8e88c4d95c")
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 int, ndim 0, dtype family int, 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

# Construct a curriculum dict with integer keys and metadata
curriculum = {
    0: {
        'img_size': 64,
        'learning_rate': 0.001,
    },
    10: {
        'img_size': 128,
        'learning_rate': 0.0005,
    },
    20: {
        'img_size': 256,
        'learning_rate': 0.0001,
    },
    'constant_param': 'some_value',
    'another_constant': 42,
}

current_step = 15

ARGS = {
    'curriculum': curriculum,
    'current_step': current_step,
}
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 int, ndim 0, dtype family int, 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
next_upsample_step Ran
ran on a synthesized input
sheldontsui/gof_neurips2021/curriculums.py
code served (permissive licence) · get_code("738ade559b76771c")
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 int, ndim 0, dtype family int, 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

# Construct a curriculum dictionary that the function expects
curriculum = {
    # Integer keys represent curriculum steps with metadata
    0: {
        'img_size': 64,
        'other_param': 'value0'
    },
    10: {
        'img_size': 128,
        'other_param': 'value10'
    },
    20: {
        'img_size': 256,
        'other_param': 'value20'
    },
    30: {
        'img_size': 512,
        'other_param': 'value30'
    },
    # Non-integer keys are global metadata
    'global_setting': 'shared_value',
    'model_type': 'test_model'
}

# current_step should be an integer representing the current training step
current_step = 15

ARGS = {
    'curriculum': curriculum,
    'current_step': current_step
}
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 int, ndim 0, dtype family int, 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
get_dataset Not yet run
unverified (harvested, no recorded run)
sheldontsui/gof_neurips2021/datasets.py
code served (permissive licence) · get_code("5d4213f664c985e6")
How it ran

no invocation recorded.

get_dataset_distributed Not yet run
unverified (harvested, no recorded run)
sheldontsui/gof_neurips2021/datasets.py
code served (permissive licence) · get_code("7db65af4855ef149")
How it ran

no invocation recorded.

sample_generator Not yet run
unverified (harvested, no recorded run)
sheldontsui/gof_neurips2021/extract_shapes.py
code served (permissive licence) · get_code("9d5c1360cf536467")
How it ran

no invocation recorded.

setup_evaluation Not yet run
unverified (harvested, no recorded run)
sheldontsui/gof_neurips2021/fid_evaluation.py
code served (permissive licence) · get_code("7e8445baa840a803")
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, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 4 later) and 6 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
create_samples 5 other papers
2 dated earlier, 3 later
2110.09788 (2021-10)
CIPS-3D: A 3D-Aware Generator of GANs Based on Conditionally-Independent Pixel Synthesis
2110.15678 (2021-10)
A Shading-Guided Generative Implicit Model for Shape-Accurate 3D-Aware Image Synthesis
2211.11208 (2022-11)
Next3D: Generative Neural Texture Rasterization for 3D-Aware Head Avatars
and 2 more not named here
get_current_step 3 other papers
2 dated earlier, 1 later
2110.09788 (2021-10)
CIPS-3D: A 3D-Aware Generator of GANs Based on Conditionally-Independent Pixel Synthesis
2110.15678 (2021-10)
A Shading-Guided Generative Implicit Model for Shape-Accurate 3D-Aware Image Synthesis
2211.11208 (2022-11)
Next3D: Generative Neural Texture Rasterization for 3D-Aware Head Avatars
last_upsample_step 3 other papers
2 dated earlier, 1 later
2110.09788 (2021-10)
CIPS-3D: A 3D-Aware Generator of GANs Based on Conditionally-Independent Pixel Synthesis
2110.15678 (2021-10)
A Shading-Guided Generative Implicit Model for Shape-Accurate 3D-Aware Image Synthesis
2211.11208 (2022-11)
Next3D: Generative Neural Texture Rasterization for 3D-Aware Head Avatars
next_upsample_step 3 other papers
2 dated earlier, 1 later
2110.09788 (2021-10)
CIPS-3D: A 3D-Aware Generator of GANs Based on Conditionally-Independent Pixel Synthesis
2110.15678 (2021-10)
A Shading-Guided Generative Implicit Model for Shape-Accurate 3D-Aware Image Synthesis
2211.11208 (2022-11)
Next3D: Generative Neural Texture Rasterization for 3D-Aware Head Avatars
get_dataset 1 other paper
1 dated earlier
2110.09788 (2021-10)
CIPS-3D: A 3D-Aware Generator of GANs Based on Conditionally-Independent Pixel Synthesis

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

The advent of generative radiance fields has significantly promoted the development of 3D-aware image synthesis. The cumulative rendering process in radiance fields makes training these generative models much easier since gradients are distributed over the entire volume, but leads to diffused object surfaces. In the meantime, compared to radiance fields occupancy representations could inherently ensure deterministic surfaces. However, if we directly apply occupancy representations to generative models, during training they will only receive sparse gradients located on object surfaces and eventually suffer from the convergence problem. In this paper, we propose Generative Occupancy Fields (GOF), a novel model based on generative radiance fields that can learn compact object surfaces without impeding its training convergence. The key insight of GOF is a dedicated transition from the cumulative rendering in radiance fields to rendering with only the surface points as the learned surface gets more and more accurate. In this way, GOF combines the merits of two representations in a unified framework. In practice, the training-time transition of start from radiance fields and march to occupancy representations is achieved in GOF by gradually shrinking the sampling region in its rendering process from the entire volume to a minimal neighboring region around the surface. Through comprehensive experiments on multiple datasets, we demonstrate that GOF can synthesize high-quality images with 3D consistency and simultaneously learn compact and smooth object surfaces. Code, models, and demo videos are available at https://sheldontsui.github.io/projects/GOF

For agents

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

get_harvested_code_for_paper("2111.00969")
get_code_for_paper("2111.00969")
have("2111.00969")

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

curl https://syntology.ai/api/ran/2111.00969.json

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

[![Syntology run record](https://syntology.ai/api/ran/2111.00969.svg)](https://syntology.ai/paper/2111.00969)

Connect an agent — have() is free.