SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2405.14832 · 2024

Direct3D: Scalable Image-to-3D Generation via 3D Latent Diffusion Transformer

arXiv · PDF · Open in the Atlas

Code that ran

5 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, 0 violated it and 4 have no contract verdict; 1 failure of our instrument; 2 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/2405.14832.json · badge: /api/ran/2405.14832.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
DreamTechAI/Direct3D community (classifier label)
classifier basis not recorded on the link
5 ran · 2 no recorded run
FunctionStatusWhere it lives
crop_recenter Ran
ran on a synthesized input
DreamTechAI/Direct3D/direct3d/utils/image.py
code served (permissive licence) · get_code("156955b54770f9fe")
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 Image. 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 PIL import Image

# Create a test image with alpha channel (RGBA)
# Image dimensions: 100x80 (height x width)
height, width = 100, 80
channels = 4

# Create numpy array with RGBA channels
image_array = np.zeros((height, width, channels), dtype=np.uint8)

# Fill RGB channels with some color
image_array[:, :, 0] = 100  # Red channel
image_array[:, :, 1] = 150  # Green channel
image_array[:, :, 2] = 200  # Blue channel

# Create alpha channel with a mask
# Add some non-transparent pixels in the center
image_array[20:80, 15:70, 3] = 255  # Opaque center region
image_array[0:20, :, 3] = 50  # Semi-transparent top
image_array[80:, :, 3] = 50  # Semi-transparent bottom

# Convert to PIL Image
image_pil = Image.fromarray(image_array, mode='RGBA')

ARGS = {
    'image_no_bg': image_pil,
    'thereshold': 100
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
get_obj_from_str Ran
ran; our contract draft was wrong, not the code
DreamTechAI/Direct3D/direct3d/utils/util.py
code served (permissive licence) · get_code("221b2d116fdf1032")
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) and returned type. 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 importlib
import torch
import torch.nn as nn

torch.manual_seed(0)

# The function get_obj_from_str takes a string like "module.path.ClassName"
# and returns the class/object by:
# 1. Splitting on the last dot to get module path and class name
# 2. Optionally reloading the module if reload=True
# 3. Importing the module and getting the attribute (class/object)

# We need to provide:
# - string: a valid module.class path that exists
# - reload: boolean flag

# Using a built-in class that's guaranteed to exist
string_value = "torch.nn.Linear"
reload_value = False

ARGS = {
    "string": string_value,
    "reload": reload_value,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type type — a type output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a type 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
pad_to_same_size Ran
ran on a synthesized input
DreamTechAI/Direct3D/direct3d/utils/image.py
code served (permissive licence) · get_code("ff87ca58669d9c48")
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 Image. 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 PIL import Image

# Create a rectangular image (non-square) with RGB channels
# Using distinct dimensions: 4 height, 6 width
image_rgb = np.random.randint(0, 256, size=(4, 6, 3), dtype=np.uint8)

# Wrap it in a PIL Image
pil_image = Image.fromarray(image_rgb)

ARGS = {
    "image": pil_image,
    "pad_value": 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
project_onto_planes Ran
ran on a synthesized input
DreamTechAI/Direct3D/direct3d/utils/triplane.py
code served (permissive licence) · get_code("fbf30db516b62ca9")
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 3, dtype family float, shape (6, 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 torch

torch.manual_seed(0)

# Create planes: shape (n_planes, 3, 3)
# Each plane is defined by 3 basis vectors (3x3 matrix)
n_planes = 2
planes = torch.randn(n_planes, 3, 3)

# Create coordinates: shape (N, M, 3)
# N = batch size, M = number of points, 3 = xyz coordinates
N = 3
M = 4
coordinates = torch.randn(N, M, 3)

ARGS = {
    "planes": planes,
    "coordinates": coordinates,
}
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 3, dtype family float, shape (6, 4, 2) — shape (6, 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
sample_from_planes Ran
ran on a synthesized input
DreamTechAI/Direct3D/direct3d/utils/triplane.py
code served (permissive licence) · get_code("4a3ead4396ff6969")
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, 3, 5, 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 torch.nn as nn
import torch.nn.functional as F
from einops import rearrange

torch.manual_seed(0)

# Create small plane axes: shape (n_planes=3, 3, 3)
# Each plane is defined by 3 basis vectors of size 3
plane_axes = torch.randn(3, 3, 3, dtype=torch.float32)

# Create plane features: shape (N_b=2, N_t=3, C=4, H_t=8, W_t=8)
# N_b is batch size, N_t is number of planes, C is channels, H_t and W_t are spatial dims
plane_features = torch.randn(2, 3, 4, 8, 8, dtype=torch.float32)

# Create coordinates: shape (N=2, M=5, 3)
# N is batch size, M is number of points, 3 is xyz coordinates
coordinates = torch.randn(2, 5, 3, dtype=torch.float32)

# Mode for grid_sample
mode = 'bilinear'

# Padding mode for grid_sample
padding_mode = 'zeros'

# Box warp scaling factor
box_warp = 2.0

ARGS = {
    'plane_axes': plane_axes,
    'plane_features': plane_features,
    'coordinates': coordinates,
    'mode': mode,
    'padding_mode': padding_mode,
    'box_warp': box_warp,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, 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, 3, 5, 4) — shape (2, 3, 5, 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_2d_sincos_pos_embed Not yet run
unverified (harvested, no recorded run)
DreamTechAI/Direct3D/direct3d/models/dit.py
code served (permissive licence) · get_code("2f540c03b489c19d")
How it ran

no invocation recorded.

instantiate_from_config Not yet run
unverified (harvested, no recorded run)
DreamTechAI/Direct3D/direct3d/utils/util.py
code served (permissive licence) · get_code("ef52d142b2c30ac2")
How it ran

no invocation recorded.

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, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 2 the same month, 2 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_obj_from_str 54 other papers
30 dated earlier, 1 the same month, 22 later, 1 same year, order not determinable · 12 only as an assembly brick
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2102.05379 (2021-02)
Argmax Flows and Multinomial Diffusion: Learning Categorical Distributions
2112.00384 (2021-12)
Exploration into Translation-Equivariant Image Quantization
and 51 more not named here
instantiate_from_config 24 other papers
17 dated earlier, 1 the same month, 6 later · 7 only as an assembly brick
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2102.05379 (2021-02)
Argmax Flows and Multinomial Diffusion: Learning Categorical Distributions
2206.02262 (2022-06)
Diffusion-GAN: Training GANs with Diffusion
and 21 more not named here

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

Generating high-quality 3D assets from text and images has long been challenging, primarily due to the absence of scalable 3D representations capable of capturing intricate geometry distributions. In this work, we introduce Direct3D, a native 3D generative model scalable to in-the-wild input images, without requiring a multiview diffusion model or SDS optimization. Our approach comprises two primary components: a Direct 3D Variational Auto-Encoder (D3D-VAE) and a Direct 3D Diffusion Transformer (D3D-DiT). D3D-VAE efficiently encodes high-resolution 3D shapes into a compact and continuous latent triplane space. Notably, our method directly supervises the decoded geometry using a semi-continuous surface sampling strategy, diverging from previous methods relying on rendered images as supervision signals. D3D-DiT models the distribution of encoded 3D latents and is specifically designed to fuse positional information from the three feature maps of the triplane latent, enabling a native 3D generative model scalable to large-scale 3D datasets. Additionally, we introduce an innovative image-to-3D generation pipeline incorporating semantic and pixel-level image conditions, allowing the model to produce 3D shapes consistent with the provided conditional image input. Extensive experiments demonstrate the superiority of our large-scale pre-trained Direct3D over previous image-to-3D approaches, achieving significantly better generation quality and generalization ability, thus establishing a new state-of-the-art for 3D content creation. Project page: https://nju-3dv.github.io/projects/Direct3D/.

For agents

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

get_harvested_code_for_paper("2405.14832")
get_code_for_paper("2405.14832")
have("2405.14832")

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

curl https://syntology.ai/api/ran/2405.14832.json

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

[![Syntology run record](https://syntology.ai/api/ran/2405.14832.svg)](https://syntology.ai/paper/2405.14832)

Connect an agent — have() is free.