SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2312.09168 · 2023

DiffusionLight: Light Probes for Free by Painting a Chrome Ball

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; 2 honoured the contract we drafted, 2 violated it and 0 have no contract verdict; 2 failures 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/2312.09168.json · badge: /api/ran/2312.09168.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
DiffusionLight/DiffusionLight canonical (classifier label)
classifier basis: classifier_2025
6 ran · 0 no recorded run
FunctionStatusWhere it lives
create_envmap_grid Ran
ran, honoured the contract we drafted
DiffusionLight/DiffusionLight/ball2envmap.py
code served (permissive licence) · get_code("514c2ce742c55045")
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 (8, 16, 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 numpy as np
import torch

torch.manual_seed(0)

ARGS = {
    "size": 8
}
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 ndarray, ndim 3, dtype family float, shape (8, 16, 2) — shape (8, 16, 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
get_ball_location Ran
ran, honoured the contract we drafted
DiffusionLight/DiffusionLight/inpaint.py
code served (permissive licence) · get_code("ce4effc348048495")
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 int; 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
from types import SimpleNamespace

torch.manual_seed(0)

# Construct image_data with boundary information
image_data = {
    'boundary': {
        'x': 100,
        'y': 150,
        'size': 25
    }
}

# Construct args as a SimpleNamespace with all required attributes
args = SimpleNamespace(
    ball_dilate=10,
    img_width=640,
    img_height=480,
    ball_size=20
)

ARGS = {
    'image_data': image_data,
    'args': args
}
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, ndim 1, dtype family int — 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_cartesian_from_spherical Ran
ran, violated the contract we drafted
DiffusionLight/DiffusionLight/ball2envmap.py
code served (permissive licence) · get_code("b643137afa71c998")
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 2, dtype family float, shape (2, 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 numpy as np

# Create sample spherical coordinates
theta = np.array([0.5, 1.2])  # vertical angles
phi = np.array([0.3, 2.1])    # horizontal angles
r = 1.0                         # radius

ARGS = {
    "theta": theta,
    "phi": phi,
    "r": r
}
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 2, dtype family float, shape (2, 3) — shape (2, 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
get_normal_vector Ran
ran, violated the contract we drafted
DiffusionLight/DiffusionLight/ball2envmap.py
code served (permissive licence) · get_code("360bcc2d023ca2dd")
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 ndarray, ndim 2, dtype family float, shape (2, 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 numpy as np

# Create sample incoming and reflect vectors
# Shape: (batch_size, vector_dim) where batch_size=2, vector_dim=3
incoming_vector = np.array([
    [1.0, 0.0, 0.0],
    [0.0, 1.0, 0.0]
], dtype=np.float32)

reflect_vector = np.array([
    [0.0, 1.0, 0.0],
    [1.0, 0.0, 0.0]
], dtype=np.float32)

ARGS = {
    "incoming_vector": incoming_vector,
    "reflect_vector": reflect_vector
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 2, dtype family float, shape (2, 3) — shape (2, 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
interpolate_embedding Ran
ran; our contract draft was wrong, not the code
DiffusionLight/DiffusionLight/inpaint.py
code served (permissive licence) · get_code("b6cfcee4c507bb23")
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 dict. 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 torch.nn as nn

torch.manual_seed(0)

# Create a minimal pipeline object that has the encode_prompt method
class MockPipeline(nn.Module):
    def __init__(self):
        super().__init__()
    
    def encode_prompt(self, prompt):
        # Returns: (prompt_embeds, _, pooled_prompt_embeds, _)
        # prompt_embeds shape: (batch_size, seq_len, embedding_dim)
        # pooled_prompt_embeds shape: (batch_size, pooled_dim)
        batch_size = 1
        seq_len = 7
        embedding_dim = 4
        pooled_dim = 5
        
        prompt_embeds = torch.randn(batch_size, seq_len, embedding_dim)
        pooled_prompt_embeds = torch.randn(batch_size, pooled_dim)
        
        return prompt_embeds, None, pooled_prompt_embeds, None

class MockPipe:
    def __init__(self):
        self.pipeline = MockPipeline()

# Create arguments object
class Args:
    def __init__(self):
        self.ev = "0.0,0.5,1.0"
        self.max_negative_ev = 1.0
        self.prompt = "a beautiful landscape"
        self.prompt_dark = "a dark, gloomy landscape"

pipe = MockPipe()
args = Args()

ARGS = {
    "pipe": pipe,
    "args": args
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
parse_filename Ran
ran; our contract draft was wrong, not the code
DiffusionLight/DiffusionLight/exposure2hdr.py
code served (permissive licence) · get_code("da6bb4fd0e3ea527")
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 dict. 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)

# Analyzing the function:
# - ev_string: used in filename.split(ev_string) and ev_string.join(a[:-1])
#   This should be a string separator
# - endwith: used in a[-1].replace(endwith, '')
#   This should be a string suffix to remove
# - filename: used in filename.split(ev_string) and returned
#   This should be a string containing ev_string and endwith

# Constructing realistic test arguments
ev_string = "_EV"
endwith = ".jpg"
filename = "photo_001_EV50.jpg"

# The function will:
# 1. Split "photo_001_EV50.jpg" by "_EV" -> ["photo_001", "50.jpg"]
# 2. Join everything except last with "_EV" -> "photo_001"
# 3. Remove ".jpg" from "50.jpg" -> "50"
# 4. Convert to int and divide by 10 -> 5.0

ARGS = {
    'ev_string': ev_string,
    'endwith': endwith,
    'filename': filename
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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

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, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 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.

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 simple yet effective technique to estimate lighting in a single input image. Current techniques rely heavily on HDR panorama datasets to train neural networks to regress an input with limited field-of-view to a full environment map. However, these approaches often struggle with real-world, uncontrolled settings due to the limited diversity and size of their datasets. To address this problem, we leverage diffusion models trained on billions of standard images to render a chrome ball into the input image. Despite its simplicity, this task remains challenging: the diffusion models often insert incorrect or inconsistent objects and cannot readily generate images in HDR format. Our research uncovers a surprising relationship between the appearance of chrome balls and the initial diffusion noise map, which we utilize to consistently generate high-quality chrome balls. We further fine-tune an LDR diffusion model (Stable Diffusion XL) with LoRA, enabling it to perform exposure bracketing for HDR light estimation. Our method produces convincing light estimates across diverse settings and demonstrates superior generalization to in-the-wild scenarios.

For agents

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

get_harvested_code_for_paper("2312.09168")
get_code_for_paper("2312.09168")
have("2312.09168")

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

curl https://syntology.ai/api/ran/2312.09168.json

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

[![Syntology run record](https://syntology.ai/api/ran/2312.09168.svg)](https://syntology.ai/paper/2312.09168)

Connect an agent — have() is free.