Papers › Signature and Log-signature for the Study of Empirical Distributions Generated with GANs

Signature and Log-signature for the Study of Empirical Distributions Generated with GANs

7 Mar 2022arXiv:2203.03226archive 2025-07-28

Joaquim de Curtò, Irene de Zarzà, Hong Yan, Carlos T. Calafate

In this paper, we bring forward the use of the recently developed Signature Transform as a way to measure the similarity between image distributions and provide detailed acquaintance and extensive evaluations. We are the first to pioneer RMSE and MAE Signature, along with log-signature as an alternative to measure GAN convergence, a problem that has been extensively studied. We are also forerunners to introduce analytical measures based on statistics to study the goodness of fit of the GAN sample distribution that are both efficient and effective. Current GAN measures involve lots of computation normally done at the GPU and are very time consuming. In contrast, we diminish the computation time to the order of seconds and computation is done at the CPU achieving the same level of goodness. Lastly, a PCA adaptive t-SNE approach, which is novel in this context, is also proposed for data visualization.

PaperPDFCode

Code

decurtoydiaz/signatures officialmentioned in papermentioned on GitHub 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

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Data VisualizationImage GenerationInstance SegmentationSemantic Segmentation

2 archive task tags without a task page not shown.

Datasets

Introduced by this paper, per the archive.

NASA Perseverance

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Generation AFHQ Cat Stylegan2-ada (NVIDIA pre-trained) MAE Signature 45968 #8 of 8 Archive leaderboard report
Image Generation AFHQ Cat Stylegan2-ada (NVIDIA pre-trained) MAE log-signature 22297 #8 of 8 Archive leaderboard report
Image Generation AFHQ Cat Stylegan2-ada (NVIDIA pre-trained) RMSE Signature 61450 #8 of 8 Archive leaderboard report
Image Generation AFHQ Cat Stylegan2-ada (NVIDIA pre-trained) RMSE log-signature 29201 #8 of 8 Archive leaderboard report
Image Generation AFHQ Dog Stylegan2-ada (NVIDIA pre-trained) MAE Signature 30441 #6 of 6 Archive leaderboard report
Image Generation AFHQ Dog Stylegan2-ada (NVIDIA pre-trained) MAE log-signature 24612 #6 of 6 Archive leaderboard report
Image Generation AFHQ Dog Stylegan2-ada (NVIDIA pre-trained) RMSE Signature 38861 #6 of 6 Archive leaderboard report
Image Generation AFHQ Dog Stylegan2-ada (NVIDIA pre-trained) RMSE log-signature 31686 #6 of 6 Archive leaderboard report
Image Generation AFHQ Wild Stylegan2-ada (NVIDIA pre-trained) MAE Signature 25578 #5 of 5 Archive leaderboard report
Image Generation AFHQ Wild Stylegan2-ada (NVIDIA pre-trained) MAE log-signature 20359 #5 of 5 Archive leaderboard report
Image Generation AFHQ Wild Stylegan2-ada (NVIDIA pre-trained) RMSE Signature 33306 #5 of 5 Archive leaderboard report
Image Generation AFHQ Wild Stylegan2-ada (NVIDIA pre-trained) RMSE log-signature 26622 #5 of 5 Archive leaderboard report
Image Generation MetFaces t-Stylegan3-ada (NVIDIA pre-trained) MAE Signature 19872 #1 of 3 Archive leaderboard report
Image Generation MetFaces t-Stylegan3-ada (NVIDIA pre-trained) MAE log-signature 13761 #1 of 3 Archive leaderboard report
Image Generation MetFaces t-Stylegan3-ada (NVIDIA pre-trained) RMSE Signature 30894 #1 of 3 Archive leaderboard report
Image Generation MetFaces t-Stylegan3-ada (NVIDIA pre-trained) RMSE log-signature 21560 #1 of 3 Archive leaderboard report
Image Generation MetFaces r-Stylegan3-ada (NVIDIA pre-trained) MAE Signature 22799 #2 of 3 Archive leaderboard report
Image Generation MetFaces r-Stylegan3-ada (NVIDIA pre-trained) MAE log-signature 16539 #2 of 3 Archive leaderboard report
Image Generation MetFaces r-Stylegan3-ada (NVIDIA pre-trained) RMSE Signature 34977 #2 of 3 Archive leaderboard report
Image Generation MetFaces r-Stylegan3-ada (NVIDIA pre-trained) RMSE log-signature 24707 #2 of 3 Archive leaderboard report
Image Generation MetFaces Stylegan2-ada (NVIDIA pre-trained) MAE Signature 23428 #3 of 3 Archive leaderboard report
Image Generation MetFaces Stylegan2-ada (NVIDIA pre-trained) MAE log-signature 18071 #3 of 3 Archive leaderboard report
Image Generation MetFaces Stylegan2-ada (NVIDIA pre-trained) RMSE Signature 33247 #3 of 3 Archive leaderboard report
Image Generation MetFaces Stylegan2-ada (NVIDIA pre-trained) RMSE log-signature 25685 #3 of 3 Archive leaderboard report
Image Generation NASA Perseverance Stylegan2-ada MAE Signature 9086 #1 of 1 Archive leaderboard report
Image Generation NASA Perseverance Stylegan2-ada MAE log-signature 5717 #1 of 1 Archive leaderboard report
Image Generation NASA Perseverance Stylegan2-ada RMSE Signature 11601 #1 of 1 Archive leaderboard report
Image Generation NASA Perseverance Stylegan2-ada RMSE log-signature 7397 #1 of 1 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

1x1 ConvolutionASPPBatch NormalizationDeepLabv3Dilated ConvolutionMAEPCASpatial Pyramid Pooling

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