SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.14859 · 2023

BakedSDF: Meshing Neural SDFs for Real-Time View Synthesis

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; 2 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 2 failures 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/2302.14859.json · badge: /api/ran/2302.14859.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
hugoycj/torch-bakedsdf community (classifier label)
classifier basis not recorded on the link
5 ran · 2 no recorded run
FunctionStatusWhere it lives
cast_rays Ran
ran, honoured the contract we drafted
hugoycj/torch-bakedsdf/models/ray_utils.py
code served (permissive licence) · get_code("45090c68bfc689c0")
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
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# cast_rays(ori, dir, z_vals)
# ori[..., None, :] - ori is indexed with ellipsis, None, and slice, so it's a tensor
# z_vals[..., None] - z_vals is indexed with ellipsis and None, so it's a tensor
# dir[..., None, :] - dir is indexed with ellipsis, None, and slice, so it's a tensor
# The operation: ori[..., None, :] + z_vals[..., None] * dir[..., None, :]
# This performs broadcasting of ray origins, directions, and depth values

# Let's construct tensors that will work with this broadcasting:
# ori: shape (..., 3) - ray origins in 3D space, batch size 2
# dir: shape (..., 3) - ray directions in 3D space, batch size 2  
# z_vals: shape (..., depth_samples) - depth sample values, batch size 2 with 4 depth samples

# Using distinct sizes: batch=2, spatial_dim=3, depth_samples=4
ori = torch.randn(2, 3)  # 2 rays, 3D coordinates
dir = torch.randn(2, 3)  # 2 ray directions, 3D vectors
z_vals = torch.randn(2, 4)  # 2 rays, 4 depth samples per ray

ARGS = {
    "ori": ori,
    "dir": dir,
    "z_vals": z_vals,
}
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 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
dot Ran
ran; our contract draft was wrong, not the code
hugoycj/torch-bakedsdf/models/utils.py
code served (permissive licence) · get_code("456eeade4275ae9d")
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, 1) 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 two small tensors for the dot product computation
# x and y should have compatible shapes for element-wise multiplication
# The function sums over the last dimension (-1)
x = torch.randn(2, 4)  # batch_size=2, dim=4
y = torch.randn(2, 4)  # batch_size=2, dim=4

ARGS = {
    "x": x,
    "y": y,
}
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 2, dtype family float, shape (2, 1) — shape (2, 1) 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_activation Ran
ran; our contract draft was wrong, not the code
hugoycj/torch-bakedsdf/models/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from heheyas/v3d, is pointer only (licence: NONE) · get_code("33892b8f6d142c6d")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from heheyas/v3d, and GitHub detected no licence in that repository when it was checked. 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_ray_directions Ran
ran, honoured the contract we drafted
hugoycj/torch-bakedsdf/models/ray_utils.py
code served (permissive licence) · get_code("4dd0fa230b9e8f1d")
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 (3, 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
import numpy as np

torch.manual_seed(0)

# The function takes camera intrinsics and image dimensions
# and generates ray directions for a pinhole camera model

W = 4  # image width
H = 3  # image height
fx = 500.0  # focal length x
fy = 500.0  # focal length y
cx = 2.0  # principal point x
cy = 1.5  # principal point y
use_pixel_centers = True

ARGS = {
    "W": W,
    "H": H,
    "fx": fx,
    "fy": fy,
    "cx": cx,
    "cy": cy,
    "use_pixel_centers": use_pixel_centers
}
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 3, dtype family float, shape (3, 4, 3) — shape (3, 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
quantize_and_clip Ran
ran on a synthesized input
hugoycj/torch-bakedsdf/models/texture.py
code served (permissive licence) · get_code("2c3c2b6ba2c91dbc")
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)

# The function takes 'values' and multiplies by 255
# It expects a numeric type that supports multiplication
# A tensor works perfectly for this

values = torch.randn(2, 4)

ARGS = {
    "values": values
}
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
chunk_batch Not yet run
unverified (harvested, no recorded run)
hugoycj/torch-bakedsdf/models/utils.py
code served (permissive licence) · get_code("3cde7958baaeae2e")
How it ran

no invocation recorded.

get_rays Not yet run
unverified (harvested, no recorded run)
hugoycj/torch-bakedsdf/models/ray_utils.py
code served (permissive licence) · get_code("2172b8a161d5fd31")
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, 6 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 6 later) and 1 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
dot 9 other papers
9 later
2308.07391 (2023-08)
PARIS: Part-level Reconstruction and Motion Analysis for Articulated Objects
2311.17082 (2023-11)
DreamPropeller: Supercharge Text-to-3D Generation with Parallel Sampling
2312.09527 (2023-12)
TIFace: Improving Facial Reconstruction through Tensorial Radiance Fields and Implicit Surfaces
and 6 more not named here
cast_rays 4 other papers
4 later
2308.07391 (2023-08)
PARIS: Part-level Reconstruction and Motion Analysis for Articulated Objects
2312.09527 (2023-12)
TIFace: Improving Facial Reconstruction through Tensorial Radiance Fields and Implicit Surfaces
2403.06738 (2024-03)
V3D: Video Diffusion Models are Effective 3D Generators
and 1 more not named here
chunk_batch 3 other papers
3 later
2312.09527 (2023-12)
TIFace: Improving Facial Reconstruction through Tensorial Radiance Fields and Implicit Surfaces
2403.06738 (2024-03)
V3D: Video Diffusion Models are Effective 3D Generators
2407.14257 (2024-07)
SparseCraft: Few-Shot Neural Reconstruction through Stereopsis Guided Geometric Linearization
get_activation 2 other papers
2 later
2312.09527 (2023-12)
TIFace: Improving Facial Reconstruction through Tensorial Radiance Fields and Implicit Surfaces
2403.06738 (2024-03)
V3D: Video Diffusion Models are Effective 3D Generators
get_ray_directions 2 other papers
2 later
2312.09527 (2023-12)
TIFace: Improving Facial Reconstruction through Tensorial Radiance Fields and Implicit Surfaces
2407.14257 (2024-07)
SparseCraft: Few-Shot Neural Reconstruction through Stereopsis Guided Geometric Linearization
get_rays 2 other papers
2 later
2312.09527 (2023-12)
TIFace: Improving Facial Reconstruction through Tensorial Radiance Fields and Implicit Surfaces
2407.14257 (2024-07)
SparseCraft: Few-Shot Neural Reconstruction through Stereopsis Guided Geometric Linearization

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

We present a method for reconstructing high-quality meshes of large unbounded real-world scenes suitable for photorealistic novel view synthesis. We first optimize a hybrid neural volume-surface scene representation designed to have well-behaved level sets that correspond to surfaces in the scene. We then bake this representation into a high-quality triangle mesh, which we equip with a simple and fast view-dependent appearance model based on spherical Gaussians. Finally, we optimize this baked representation to best reproduce the captured viewpoints, resulting in a model that can leverage accelerated polygon rasterization pipelines for real-time view synthesis on commodity hardware. Our approach outperforms previous scene representations for real-time rendering in terms of accuracy, speed, and power consumption, and produces high quality meshes that enable applications such as appearance editing and physical simulation.

For agents

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

get_harvested_code_for_paper("2302.14859")
get_code_for_paper("2302.14859")
have("2302.14859")

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

curl https://syntology.ai/api/ran/2302.14859.json

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

[![Syntology run record](https://syntology.ai/api/ran/2302.14859.svg)](https://syntology.ai/paper/2302.14859)

Connect an agent — have() is free.