SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1909.12573 · 2019

RGBD-GAN: Unsupervised 3D Representation Learning From Natural Image Datasets via RGBD Image Synthesis

arXiv · PDF · Open in the Atlas

Code that ran

6 of 6 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; 1 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 1 failure 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/1909.12573.json · badge: /api/ran/1909.12573.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
nogu-atsu/RGBD-GAN canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
nogu-atsu/tmp_rgbdgan community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
FunctionStatusWhere it lives
crop_square Ran
ran; our contract draft was wrong, not the code
nogu-atsu/RGBD-GAN/train_rgbd.py
pointer only (licence: NOASSERTION) · get_code("70e21270b7e47327")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, nogu-atsu/RGBD-GAN (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.

get_camera_matries Ran
ran on a synthesized input
nogu-atsu/tmp_rgbdgan/updater.py
code served (permissive licence) · get_code("2654a7e6bdc59e0c")
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, 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

# Small batch of rotation and translation parameters
# Shape: (batch_size=2, 6) where 6 = [x_rot, y_rot, z_rot, x_trans, y_trans, z_trans]
thetas = np.array([
    [0.1, 0.2, 0.3, 0.5, 0.6, 0.7],
    [0.15, 0.25, 0.35, 0.55, 0.65, 0.75]
], dtype='float32')

# Default order parameter
order = (0, 1, 2)

ARGS = {
    'thetas': thetas,
    'order': order
}
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, 4) — shape (2, 4, 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_lr_scale_factor Ran
ran on a synthesized input
nogu-atsu/tmp_rgbdgan/config.py
code served (permissive licence) · get_code("b595eb4895c8fc67")
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 math

gpu_lr = {1: {15: 1.5, 16: 1.5, 17: 1.5}, 2: {13: 1.5, 14: 1.5, 15: 2, 16: 2, 17: 2}, 3: {11: 1.5, 12: 1.5, 13: 2, 14: 2, 15: 2.5, 16: 2.5, 17: 2.5}, 4: {11: 1.5, 12: 1.5, 13: 2, 14: 2, 15: 3, 16: 3, 17: 3}, 8: {9: 1.5, 10: 1.5, 11: 2, 12: 2, 13: 3, 14: 3, 15: 3, 16: 3, 17: 3}}

def get_lr_scale_factor(total_gpu, stage):
    gpu_lr_d = gpu_lr.get(total_gpu, gpu_lr[1])
    stage = math.floor(stage)
    if stage >= 18:
        return gpu_lr_d[17]
    return gpu_lr_d.get(stage, 1)

ARGS = {
    "total_gpu": 4,
    "stage": 15.7,
}
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 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
make_dataset Ran
ran, honoured the contract we drafted
nogu-atsu/RGBD-GAN/train_rgbd.py
pointer only (licence: NOASSERTION) · get_code("78651dd7e709ddcd")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, nogu-atsu/RGBD-GAN (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.

num_divisible_by_2 Ran
ran on a synthesized input
nogu-atsu/tmp_rgbdgan/deepvoxel/deepvoxel.py
code served (permissive licence) · get_code("199023199d625be7")
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
import numpy as np

def num_divisible_by_2(number):
    i = 0
    while not number % 2:
        number = number // 2
        i += 1
    return i

# The function takes a single parameter 'number' which should be an integer
# It counts how many times the number is divisible by 2
# Test with a number divisible by 2 multiple times: 16 = 2^4
ARGS = {
    "number": 16
}
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 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
update_camera_matrices Ran
ran on a synthesized input
nogu-atsu/tmp_rgbdgan/updater.py
code served (permissive licence) · get_code("72164273ad5973fa")
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, 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

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

# Create a small minibatch of camera matrices
# The function expects mat to be a 3D array where:
# - First dimension is batch size
# - Second and third dimensions are 4x4 (homogeneous transformation matrices)
batch_size = 2
mat = np.random.randn(batch_size, 4, 4).astype(np.float32)

# axis1 and axis2 are rotation axes (0~2 for x, y, z)
axis1 = 0
axis2 = 1

# theta is an array of rotation angles in radians (one per batch element)
theta = np.array([0.5, 1.0], dtype=np.float32)

ARGS = {
    "mat": mat,
    "axis1": axis1,
    "axis2": axis2,
    "theta": theta,
}
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, 4) — shape (2, 4, 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 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 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.

FunctionAlso present inEarliest-datedOthers named
get_lr_scale_factor 1 other paper
1 dated earlier
1812.04948 (2018-12)
A Style-Based Generator Architecture for Generative Adversarial Networks

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

Understanding three-dimensional (3D) geometries from two-dimensional (2D) images without any labeled information is promising for understanding the real world without incurring annotation cost. We herein propose a novel generative model, RGBD-GAN, which achieves unsupervised 3D representation learning from 2D images. The proposed method enables camera parameter-conditional image generation and depth image generation without any 3D annotations, such as camera poses or depth. We use an explicit 3D consistency loss for two RGBD images generated from different camera parameters, in addition to the ordinal GAN objective. The loss is simple yet effective for any type of image generator such as DCGAN and StyleGAN to be conditioned on camera parameters. Through experiments, we demonstrated that the proposed method could learn 3D representations from 2D images with various generator architectures.

For agents

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

get_harvested_code_for_paper("1909.12573")
get_code_for_paper("1909.12573")
have("1909.12573")

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

curl https://syntology.ai/api/ran/1909.12573.json

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

[![Syntology run record](https://syntology.ai/api/ran/1909.12573.svg)](https://syntology.ai/paper/1909.12573)

Connect an agent — have() is free.