SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.08528 · 2023

4D Gaussian Splatting for Real-Time Dynamic Scene Rendering

arXiv · PDF · Open in the Atlas

Code that ran

6 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 failures of our instrument; 5 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/2310.08528.json · badge: /api/ran/2310.08528.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
hustvl/4DGaussians canonical (classifier label)
classifier basis: census
6 ran · 5 no recorded run
FunctionStatusWhere it lives
construct_list_of_attributes Ran
ran; our contract draft was wrong, not the code
hustvl/4DGaussians/export_perframe_3DGS.py
code served (permissive licence) · get_code("5d9c08013dc39925")
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, 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 torch
import numpy as np

torch.manual_seed(0)

# The function accesses feature_dc_shape[1] and feature_dc_shape[2]
# It accesses feature_rest_shape[1] and feature_rest_shape[2]
# It accesses scaling_shape[1]
# It accesses rotation_shape[1]
# These are all used as integers in range() calls
# So these parameters should be tuple-like or list-like with indexable dimensions

feature_dc_shape = (2, 3, 4)  # batch=2, dim1=3, dim2=4
feature_rest_shape = (2, 5, 6)  # batch=2, dim1=5, dim2=6
scaling_shape = (2, 7)  # batch=2, dim=7
rotation_shape = (2, 8)  # batch=2, dim=8

ARGS = {
    'feature_dc_shape': feature_dc_shape,
    'feature_rest_shape': feature_rest_shape,
    'scaling_shape': scaling_shape,
    'rotation_shape': rotation_shape
}
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 list, ndim 1, dtype family float — a list 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
init_3DGaussians_ply Ran
ran; our contract draft was wrong, not the code
hustvl/4DGaussians/export_perframe_3DGS.py
code served (permissive licence) · get_code("bf8835048d68fc01")
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 PlyData. 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)

# Create small tensors for the function parameters
batch_size = 2
num_points = 2

# points: (batch_size, 3) - xyz coordinates
points = torch.randn(num_points, 3)

# scales: (batch_size, 3) - scaling factors
scales = torch.randn(num_points, 3)

# rotations: (batch_size, 4) - rotation quaternions
rotations = torch.randn(num_points, 4)

# opacity: (batch_size, 1) - opacity values
opactiy = torch.randn(num_points, 1)

# shs: (batch_size, num_sh_coeffs, 3) - spherical harmonic coefficients
# feature_shape will be (1, 3) meaning first 1 coefficient is DC, rest are higher order
num_sh_coeffs = 4
shs = torch.randn(num_points, num_sh_coeffs, 3)

# feature_shape: tuple indicating (num_dc_components, num_rest_components)
# This tells us how to split the shs tensor
feature_shape = (1, 3)

ARGS = {
    'points': points,
    'scales': scales,
    'rotations': rotations,
    'opactiy': opactiy,
    'shs': shs,
    'feature_shape': feature_shape
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, plyfile, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type PlyData — a PlyData output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PlyData output the recorded ndim 2 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
l1_loss Ran
ran; our contract draft was wrong, not the code
hustvl/4DGaussians/utils/loss_utils.py
code served (permissive licence) · get_code("ac0e42d6fbcfbbe6")
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 0, dtype family float, 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)

# Create small tensors for network output and ground truth
# Both should have the same shape for the subtraction and mean operations
network_output = torch.randn(2, 4)  # batch size 2, 4 dimensions
gt = torch.randn(2, 4)  # batch size 2, 4 dimensions

ARGS = {
    "network_output": network_output,
    "gt": gt,
}
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 0, dtype family float, 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
l2_loss Ran
ran, honoured the contract we drafted
hustvl/4DGaussians/utils/loss_utils.py
code served (permissive licence) · get_code("8c3b0f873ba11813")
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 0, dtype family float, 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 torch.nn as nn

torch.manual_seed(0)

# Create small tensors for network output and ground truth
# Both should have the same shape for the loss computation
# Using batch size 2, feature dimension 4
network_output = torch.randn(2, 4)
gt = torch.randn(2, 4)

ARGS = {
    "network_output": network_output,
    "gt": gt,
}
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 0, dtype family float, 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
readImages Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
hustvl/4DGaussians/metrics.py
code served (permissive licence) · get_code("cdd00787894554b5")
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. 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 os
import tempfile
from pathlib import Path
from PIL import Image
import torch
import numpy as np

torch.manual_seed(0)

# Create temporary directories with dummy image files
temp_renders_dir = tempfile.mkdtemp()
temp_gt_dir = tempfile.mkdtemp()

renders_dir = Path(temp_renders_dir)
gt_dir = Path(temp_gt_dir)

# Create 2 small dummy images (4x3 RGB)
for i in range(2):
    # Create render image
    render_array = np.random.randint(0, 256, (3, 4, 3), dtype=np.uint8)
    render_img = Image.fromarray(render_array.transpose(1, 2, 0), mode='RGB')
    render_img.save(renders_dir / f"render_{i}.png")
    
    # Create ground truth image
    gt_array = np.random.randint(0, 256, (3, 4, 3), dtype=np.uint8)
    gt_img = Image.fromarray(gt_array.transpose(1, 2, 0), mode='RGB')
    gt_img.save(gt_dir / f"render_{i}.png")

ARGS = {
    "renders_dir": renders_dir,
    "gt_dir": gt_dir,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — 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
rotate_point_cloud Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
hustvl/4DGaussians/merge_many_4dgs.py
code served (permissive licence) · get_code("473be863d3a190d8")
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 (2, 4, 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)

# Construct arguments for rotate_point_cloud

# point_cloud: a tensor of shape (batch_size, num_points, 3) representing 3D points
point_cloud = torch.randn(2, 4, 3)

# displacement: a tensor of shape (3,) or broadcastable to add to rotated points
displacement = torch.randn(3)

# rotation_angles: a tuple of (theta, phi) representing rotation angles in radians
theta = torch.tensor(0.5)
phi = torch.tensor(0.3)
rotation_angles = (theta, phi)

# scales_bias: a scalar or tensor to scale the point cloud before rotation
scales_bias = torch.tensor(1.2)

ARGS = {
    'point_cloud': point_cloud,
    'displacement': displacement,
    'rotation_angles': rotation_angles,
    'scales_bias': scales_bias
}
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 (2, 4, 3) — shape (2, 4, 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
array_to_blob Not yet run
unverified (harvested, no recorded run)
hustvl/4DGaussians/database.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from oppo-us-research/spacetimegaussians, is pointer only (licence: NOASSERTION) · get_code("141f61e2147729f4")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from oppo-us-research/spacetimegaussians, and GitHub could not classify the licence of that repository (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. No invocation recorded.

blob_to_array Not yet run
unverified (harvested, no recorded run)
hustvl/4DGaussians/database.py
code served (permissive licence) · get_code("f77197e2eb782c27")
How it ran

no invocation recorded.

get_network Not yet run
unverified (harvested, no recorded run)
hustvl/4DGaussians/lpipsPyTorch/modules/networks.py
code served (permissive licence) · get_code("07bd0da29c4dc7bb")
How it ran

no invocation recorded.

lpips_loss Not yet run
unverified (harvested, no recorded run)
hustvl/4DGaussians/utils/loss_utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from CUHK-AIM-Group/LGS, is pointer only (licence: NOASSERTION) · get_code("00716dc6573bf4d9")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from CUHK-AIM-Group/LGS, and GitHub could not classify the licence of that repository (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. No invocation recorded.

multithread_write Not yet run
unverified (harvested, no recorded run)
hustvl/4DGaussians/render.py
code served (permissive licence) · get_code("6559d028972f58ba")
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, 7 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 6 the same month, 7 later) and 4 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
l1_loss 93 other papers
2 dated earlier, 2 the same month, 89 later
2308.04079 (2023-08)
3D Gaussian Splatting for Real-Time Radiance Field Rendering
2309.13101 (2023-09)
Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction
2310.08529 (2023-10)
GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models
and 90 more not named here
l2_loss 90 other papers
2 dated earlier, 2 the same month, 86 later
2308.04079 (2023-08)
3D Gaussian Splatting for Real-Time Radiance Field Rendering
2309.13101 (2023-09)
Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction
2310.08529 (2023-10)
GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models
and 87 more not named here
get_network 60 other papers
3 dated earlier, 1 the same month, 56 later
2203.04036 (2022-03)
StyleHEAT: One-Shot High-Resolution Editable Talking Face Generation via Pre-trained StyleGAN
2308.04079 (2023-08)
3D Gaussian Splatting for Real-Time Radiance Field Rendering
2309.13101 (2023-09)
Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction
and 57 more not named here
readImages 54 other papers
2 dated earlier, 1 the same month, 51 later
2308.04079 (2023-08)
3D Gaussian Splatting for Real-Time Radiance Field Rendering
2309.13101 (2023-09)
Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction
2310.08529 (2023-10)
GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models
and 51 more not named here
array_to_blob 4 other papers
1 the same month, 3 later
2310.10642 (2023-10)
Real-time Photorealistic Dynamic Scene Representation and Rendering with 4D Gaussian Splatting
2312.16812 (2023-12)
Spacetime Gaussian Feature Splatting for Real-Time Dynamic View Synthesis
2411.02394 (2024-11)
AutoVFX: Physically Realistic Video Editing from Natural Language Instructions
and 1 more not named here
blob_to_array 2 other papers
1 the same month, 1 later
2310.10642 (2023-10)
Real-time Photorealistic Dynamic Scene Representation and Rendering with 4D Gaussian Splatting
2411.19290 (2024-11)
TRASE: Tracking-free 4D Segmentation and Editing
lpips_loss 1 other paper
1 later
2406.16073 (2024-06)
LGS: A Light-weight 4D Gaussian Splatting for Efficient Surgical Scene Reconstruction

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

Representing and rendering dynamic scenes has been an important but challenging task. Especially, to accurately model complex motions, high efficiency is usually hard to guarantee. To achieve real-time dynamic scene rendering while also enjoying high training and storage efficiency, we propose 4D Gaussian Splatting (4D-GS) as a holistic representation for dynamic scenes rather than applying 3D-GS for each individual frame. In 4D-GS, a novel explicit representation containing both 3D Gaussians and 4D neural voxels is proposed. A decomposed neural voxel encoding algorithm inspired by HexPlane is proposed to efficiently build Gaussian features from 4D neural voxels and then a lightweight MLP is applied to predict Gaussian deformations at novel timestamps. Our 4D-GS method achieves real-time rendering under high resolutions, 82 FPS at an 800$\times$800 resolution on an RTX 3090 GPU while maintaining comparable or better quality than previous state-of-the-art methods. More demos and code are available at https://guanjunwu.github.io/4dgs/.

For agents

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

get_harvested_code_for_paper("2310.08528")
get_code_for_paper("2310.08528")
have("2310.08528")

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

curl https://syntology.ai/api/ran/2310.08528.json

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

[![Syntology run record](https://syntology.ai/api/ran/2310.08528.svg)](https://syntology.ai/paper/2310.08528)

Connect an agent — have() is free.