AI Paper Atlas

Research map / Image generation / Editing and style transfer

Guided Image-to-Image Translation With Bi-Directional Feature Transformation

Badour AlBahar, Jia-Bin Huang

ICCV 2019 · Image generation › Editing and style transfer · approach: Image-to-image translation · 114 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add bi-directional feature transformation instead of input concatenation (ablation)guided pix2pix (ResNet) · NYU v2 depth upsampling x16RMSE (cm): 11.86 -> 9.01Improved
add bi-directional feature transformation instead of input concatenation (ablation)guided pix2pix (ResNet) · DeepFashion pose transferFID: 42.330 -> 13.240Improved
make feature transformation bi-directional instead of uni-directional (ablation)FT model · NYU v2 depth upsampling x16RMSE (cm): 11.41 -> 9.01Improved
make feature transformation bi-directional instead of uni-directional (ablation)FT model · DeepFashion pose transferFID: 13.988 -> 13.240Improved
make feature transformation bi-directional instead of uni-directional (ablation)FT model · Texture transfer, ShoesFID: 119.588 -> 119.832No clear effect
use 4 FT layers instead of 1 in bFT (ablation)bFT · NYU v2 depth upsampling x16RMSE (cm): 10.79 -> 9.01Improved
use 4 FT layers instead of 3 in bFT (ablation)bFT · NYU v2 depth upsampling x16RMSE (cm): 8.82 -> 9.01Hurt
use 3 FT layers instead of 1 in uFT pose transfer (ablation)uFT · DeepFashion pose transferFID: 59.678 -> 16.069Improved
use FT instead of AdaIN in bi-directional conditioning (ablation)bFT · NYU v2 depth upsampling x16RMSE (cm): 13.36 -> 9.01Improved
use FT in every layer instead of only the final layer (ablation)bFT · NYU v2 depth upsampling x16RMSE (cm): 11.40 -> 9.01Improved
apply bFT between input encoder and guide encoder instead of decoder-decoder (ablation)bFT · NYU v2 depth upsampling x16RMSE (cm): 10.97 -> 9.01Improved
apply bFT between input encoder and guide encoder instead of encoder-decoder (ablation)bFT · NYU v2 depth upsampling x16RMSE (cm): 11.96 -> 9.01Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy