Papers › Fine-Tuning Image-Conditional Diffusion Models is Easier than You Think

Fine-Tuning Image-Conditional Diffusion Models is Easier than You Think

17 Sep 2024arXiv:2409.11355archive 2025-07-28

Gonzalo Martin Garcia, Karim Abou Zeid, Christian Schmidt, Daan de Geus, Alexander Hermans, Bastian Leibe

Recent work showed that large diffusion models can be reused as highly precise monocular depth estimators by casting depth estimation as an image-conditional image generation task. While the proposed model achieved state-of-the-art results, high computational demands due to multi-step inference limited its use in many scenarios. In this paper, we show that the perceived inefficiency was caused by a flaw in the inference pipeline that has so far gone unnoticed. The fixed model performs comparably to the best previously reported configuration while being more than 200× faster. To optimize for downstream task performance, we perform end-to-end fine-tuning on top of the single-step model with task-specific losses and get a deterministic model that outperforms all other diffusion-based depth and normal estimation models on common zero-shot benchmarks. We surprisingly find that this fine-tuning protocol also works directly on Stable Diffusion and achieves comparable performance to current state-of-the-art diffusion-based depth and normal estimation models, calling into question some of the conclusions drawn from prior works.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 17 code samples harvested from 1 repository linked to this paper: 16 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 3 honoured, 2 violated, 6 with no contract checked; 5 where Syntology's instrument failed) · 1 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 17 samples from 1 repository, 16 ran (11 with no instrument failure; 5 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

VisualComputingInstitute/diffusion-e2e-ft officialmentioned on GitHubpytorchSyntology: 17 samples, 16 ran (5 instrument-failed) · 1 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

17 samples harvested: 16 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 3 honoured, 2 violated, 6 with no contract checked; 5 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · honoured contract
2ran · violated contract
3ran · our draft was wrong
2ran; the contract check could not run
6ran
1unverified

Licence: 17 of the 17 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from VisualComputingInstitute/diffusion-e2e-ft. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

VisualComputingInstitute/diffusion-e2e-ft official repository 17 samples, 16 ran (5 instrument-failed) · 1 unverified
compute_scale_and_shift_masked training/util/loss.py ran no licence file found · pointer only · 3b234f415dac393b · report fixture: signature prediction:Tensor/3/float, target:Tensor/3/float, mask:Tensor/3/float · returned tuple · level recorded 2026-09-24
define_loss DSINE/projects/baseline_normal/losses.py ran · our draft was wrong no licence file found · pointer only · b48c284df8bc63a7 · report fixture: signature loss_name:str/0/str · returned function · level recorded 2026-09-24
get_r_pitch DSINE/utils/rotation.py ran · honoured contract fingerprinted no licence file found · pointer only · 2c816eedd2f0474c · report fixture: signature pitch:float/0/float · returned ndarray · level recorded 2026-09-24
get_r_roll DSINE/utils/rotation.py ran · honoured contract fingerprinted no licence file found · pointer only · 85d2c7386475ae37 · report fixture: signature roll:float/0/float · returned ndarray · level recorded 2026-09-24
get_r_yaw DSINE/utils/rotation.py ran · honoured contract fingerprinted no licence file found · pointer only · 14244d03bf3e5f48 · report fixture: signature yaw:float/0/float · returned ndarray · level recorded 2026-09-24
intrins_crop DSINE/utils/projection.py ran fingerprinted no licence file found · pointer only · 07d8c13c1190a20a · report fixture: signature intrins:Tensor/4/float, crop_top:int/0/int, crop_left:int/0/int · returned Tensor · level recorded 2026-09-24
intrins_to_zero DSINE/utils/projection.py ran · violated contract fingerprinted no licence file found · pointer only · 1218c2650fe6ffed · report fixture: signature intrins:Tensor/3/float · returned Tensor · level recorded 2026-09-24
intrins_zero_to DSINE/utils/projection.py ran · violated contract fingerprinted no licence file found · pointer only · f56119d671b23caa · report fixture: signature intrins:Tensor/3/float · returned Tensor · level recorded 2026-09-24
l1_loss in 149 papers
census: 3 implementations from 7 papers, 3 distinct outputs across 3 buckets, one shared input per bucket
DSINE/projects/baseline_normal/losses.py ran no licence file found · pointer only · 30d6db04f2c86d1d · report fixture: signature norm_out:Tensor/3/float, gt_norm:Tensor/3/float, gt_norm_mask:Tensor/3/bool · returned Tensor · level recorded 2026-09-24
l2_loss in 136 papers DSINE/projects/baseline_normal/losses.py ran no licence file found · pointer only · 542e39255f62d87e · report fixture: signature norm_out:Tensor/3/float, gt_norm:Tensor/3/float, gt_norm_mask:Tensor/3/bool · returned Tensor · level recorded 2026-09-24
load_checkpoint in 277 papers DSINE/utils/utils.py ran no licence file found · pointer only · 769dc6d3c6aa40de · report fixture: signature fpath:str/0/str, model:callable/0/? · returned SimpleModel · level recorded 2026-09-24
pyramid_noise_like GeoWizard/geowizard/models/geowizard_pipeline.py ran no licence file found · pointer only · 11eeeb13d2957ebf · report fixture: signature x:Tensor/4/float, timesteps:Tensor/1/float, discount:float/0/float · returned Tensor · level recorded 2026-09-24
setup_custom_logger in 6 papers DSINE/utils/utils.py ran · our draft was wrong no licence file found · pointer only · dc4ca4a94df3d89e · report fixture: signature name:str/0/str, test:bool/0/bool · returned Logger · level recorded 2026-09-24
tensor_to_numpy in 10 papers DSINE/utils/visualize.py ran; the contract check could not run (RAISES) no licence file found · pointer only · e4738f6dce471506 · report fixture: signature tensor_in:Tensor/3/float · returned ndarray · level recorded 2026-09-24
txt_to_list DSINE/utils/utils.py ran · our draft was wrong no licence file found · pointer only · 2324fc8ae05d9f6d · report fixture: signature txt_path:str/0/str · returned list · level recorded 2026-09-24
unnormalize in 21 papers DSINE/utils/visualize.py ran; the contract check could not run (RAISES) no licence file found · pointer only · 61cf14e229883558 · report fixture: signature img_in:Tensor/4/float, img_stats:dict/0/? · returned ndarray · level recorded 2026-09-24
normal_to_rgb DSINE/utils/visualize.py unverified no licence file found · pointer only · 774236d466c851a3 · report

Tasks

Conditional Image GenerationDepth EstimationImage GenerationMonocular Depth EstimationSurface Normal EstimationSurface Normals Estimation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Monocular Depth Estimation NYU-Depth V2 Marigold + E2E FT(zero-shot) Delta < 1.25 0.966 #9 of 85 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 Marigold + E2E FT(zero-shot) absolute relative error 0.052 #9 of 85 Archive leaderboard report
Surface Normals Estimation IBims-1 Marigold + E2E FT(zero-shot) % < 11.25 69.9 #1 of 2 Archive leaderboard report
Surface Normals Estimation IBims-1 Marigold + E2E FT(zero-shot) Mean 15.8 #1 of 2 Archive leaderboard report
Surface Normals Estimation NYU Depth v2 Marigold + E2E FT(zero-shot) % < 11.25 61.4 #5 of 6 Archive leaderboard report
Surface Normals Estimation NYU Depth v2 Marigold + E2E FT(zero-shot) Mean Angle Error 16.2 #5 of 6 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Diffusion

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections