Papers › Image Shape Manipulation from a Single Augmented Training Sample

Image Shape Manipulation from a Single Augmented Training Sample

13 Sep 2021ICCV 2021 10arXiv:2109.06151archive 2025-07-28

Yael Vinker, Eliahu Horwitz, Nir Zabari, Yedid Hoshen

In this paper, we present DeepSIM, a generative model for conditional image manipulation based on a single image. We find that extensive augmentation is key for enabling single image training, and incorporate the use of thin-plate-spline (TPS) as an effective augmentation. Our network learns to map between a primitive representation of the image to the image itself. The choice of a primitive representation has an impact on the ease and expressiveness of the manipulations and can be automatic (e.g. edges), manual (e.g. segmentation) or hybrid such as edges on top of segmentations. At manipulation time, our generator allows for making complex image changes by modifying the primitive input representation and mapping it through the network. Our method is shown to achieve remarkable performance on image manipulation tasks.

PaperPDFConference PDFCode

Code

eliahuhorwitz/DeepSIM officialmentioned on GitHubpytorchNOASSERTION 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

Image GenerationImage ManipulationImage-to-Image TranslationSketch-to-Image Translation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Manipulation LRS2 TPS LPIPS (S1) 0.12 #1 of 2 Archive leaderboard report
Image Manipulation LRS2 TPS LPIPS (S2) 0.21 #1 of 2 Archive leaderboard report
Image Manipulation LRS2 TPS LPIPS (S3) 0.1 #1 of 2 Archive leaderboard report
Image Manipulation LRS2 TPS LPIPS (S4) 0.22 #1 of 2 Archive leaderboard report
Image Manipulation LRS2 TPS LPIPS (S5) 0.14 #1 of 2 Archive leaderboard report
Image Manipulation LRS2 TPS SIFID (S1) 0.07 #1 of 2 Archive leaderboard report
Image Manipulation LRS2 TPS SIFID (S2) 0.12 #1 of 2 Archive leaderboard report
Image Manipulation LRS2 TPS SIFID (S3) 0.04 #1 of 2 Archive leaderboard report
Image Manipulation LRS2 TPS SIFID (S4) 0.12 #1 of 2 Archive leaderboard report
Image Manipulation LRS2 TPS SIFID (S5) 0.06 #1 of 2 Archive leaderboard report
Image Manipulation LRS2 Pix2PixHD-SIA LPIPS (S1) 0.44 #2 of 2 Archive leaderboard report
Image Manipulation LRS2 Pix2PixHD-SIA LPIPS (S2) 0.47 #2 of 2 Archive leaderboard report
Image Manipulation LRS2 Pix2PixHD-SIA LPIPS (S3) 0.41 #2 of 2 Archive leaderboard report
Image Manipulation LRS2 Pix2PixHD-SIA LPIPS (S4) 0.53 #2 of 2 Archive leaderboard report
Image Manipulation LRS2 Pix2PixHD-SIA LPIPS (S5) 0.46 #2 of 2 Archive leaderboard report
Image Manipulation LRS2 Pix2PixHD-SIA SIFID (S1) 0.51 #2 of 2 Archive leaderboard report
Image Manipulation LRS2 Pix2PixHD-SIA SIFID (S2) 0.49 #2 of 2 Archive leaderboard report
Image Manipulation LRS2 Pix2PixHD-SIA SIFID (S3) 0.5 #2 of 2 Archive leaderboard report
Image Manipulation LRS2 Pix2PixHD-SIA SIFID (S4) 0.26 #2 of 2 Archive leaderboard report
Image Manipulation LRS2 Pix2PixHD-SIA SIFID (S5) 0.44 #2 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

DeepSIM

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