Papers › Palette: Image-to-Image Diffusion Models

Palette: Image-to-Image Diffusion Models

10 Nov 2021arXiv:2111.05826archive 2025-07-28

Chitwan Saharia, William Chan, Huiwen Chang, Chris A. Lee, Jonathan Ho, Tim Salimans, David J. Fleet, Mohammad Norouzi

This paper develops a unified framework for image-to-image translation based on conditional diffusion models and evaluates this framework on four challenging image-to-image translation tasks, namely colorization, inpainting, uncropping, and JPEG restoration. Our simple implementation of image-to-image diffusion models outperforms strong GAN and regression baselines on all tasks, without task-specific hyper-parameter tuning, architecture customization, or any auxiliary loss or sophisticated new techniques needed. We uncover the impact of an L2 vs. L1 loss in the denoising diffusion objective on sample diversity, and demonstrate the importance of self-attention in the neural architecture through empirical studies. Importantly, we advocate a unified evaluation protocol based on ImageNet, with human evaluation and sample quality scores (FID, Inception Score, Classification Accuracy of a pre-trained ResNet-50, and Perceptual Distance against original images). We expect this standardized evaluation protocol to play a role in advancing image-to-image translation research. Finally, we show that a generalist, multi-task diffusion model performs as well or better than task-specific specialist counterparts. Check out https://diffusion-palette.github.io for an overview of the results.

PaperPDFConference PDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2111.05826")

Code

Syntology Ran 2 of 3 code samples harvested from 1 repository linked to this paper; 1 has no recorded run. Of those that ran: 1 ran · honoured contract; 1 ran · fixture could not drive it.

By repository: community (archive-listed): 2 samples from 1 repository, 2 ran; 1 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

crosszamirski/guided-i2i mentioned on GitHubpytorchApache-2.0 report
kylelo/roofdiffusion mentioned on GitHubpytorch report
omerb01/puq mentioned on GitHubpytorch 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

3 samples harvested; 2 ran; 1 honoured the contract we drafted; 1 has no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · honoured contract
1ran · fixture could not drive it
1unverified

Licence: 3 of the 3 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 kylelo/roofdiffusion. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

IoU kylelo/roofdiffusion/data/util/roof_metric.py community (archive-listed) ran · honoured contract fingerprinted no licence file found · pointer only · ec2d227dfc59af9d · report
calculate_metrics kylelo/roofdiffusion/data/util/roof_metric.py community (archive-listed) ran · fixture could not drive it fingerprinted no licence file found · pointer only · c9f8bc1be2cb6a09 · report
extract identical code first harvested elsewhere unverified licence of this copy not recorded · 09c8479d9a5b3e06 · report

Tasks

ColorizationDenoisingDiversityImage-to-Image TranslationJPEG DecompressionPerceptual DistanceTranslationUncropping

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Colorization ImageNet ctest10k Palette FID 3.4 #1 of 1 Archive leaderboard report
Colorization ImageNet val Palette FID-5K 15.78 #1 of 4 Archive leaderboard report
Image Inpainting Places2 val Palatte (20-30% free form) FID 11.7 #1 of 7 Archive leaderboard report
Image Inpainting Places2 val Palatte (20-30% free form) PD 35.0 #1 of 7 Archive leaderboard report
Image Inpainting Places2 val Palette (128×128 center mask) FID 11.9 #2 of 7 Archive leaderboard report
Image Inpainting Places2 val Palette (128×128 center mask) PD 57.3 #2 of 7 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 20) CA 73.5 #1 of 6 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 20) FID-5K 4.3 #1 of 6 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 20) IS 208.7 #1 of 6 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 20) PD 37.1 #1 of 6 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 10) CA 70.7 #2 of 6 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 10) FID-5K 5.4 #2 of 6 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 10) IS 180.5 #2 of 6 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 10) PD 58.3 #2 of 6 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 5) CA 64.2 #3 of 6 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 5) FID-5K 8.3 #3 of 6 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 5) IS 133.6 #3 of 6 Archive leaderboard report
JPEG Decompression ImageNet Palette (QF: 5) PD 95.5 #3 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 20) CA 69.7 #4 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 20) FID-5K 11.5 #4 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 20) IS 158.7 #4 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 20) PD 65.4 #4 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 10) CA 63.5 #5 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 10) FID-5K 18.0 #5 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 10) IS 117.2 #5 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 10) PD 102.2 #5 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 5) CA 52.8 #6 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 5) FID-5K 29.0 #6 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 5) IS 73.9 #6 of 6 Archive leaderboard report
JPEG Decompression ImageNet Regression (QF: 5) PD 155.4 #6 of 6 Archive leaderboard report
Uncropping Places2 val Palette FID 3.53 #1 of 2 Archive leaderboard report
Uncropping Places2 val Palette Fool rate 39.9 #1 of 2 Archive leaderboard report
Uncropping Places2 val Palette PD 103.3 #1 of 2 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