AI Paper Atlas

Research map / Image generation / Editing and style transfer

One-Shot Neural Face Reenactment via Finding Directions in GAN’s Latent Space

Stella Bounareli, Christos Tzelepis, Vasileios Argyriou, Ioannis Patras, Georgios Tzimiropoulos

IJCV 2024 · Image generation › Editing and style transfer · approach: Identity-preserving face manipulation · 13 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 generator weight optimization at inference (ablation)LOR · VoxCeleb1 self-reenactmentCSIM: 0.37 -> 0.66Improved
add joint training of inversion encoder and directions (no optimization) (ablation)LOR+ · VoxCeleb1 self-reenactmentARD: 1.1 -> 0.8Improved
add feature space refinement (FSR) (ablation)LOR+ · VoxCeleb1 self-reenactmentARD: 0.8 -> 0.7Improved
add feature space refinement (FSR) (ablation)LOR+ · VoxCeleb1 self-reenactment, large head pose (Benchmark-L)CSIM: 0.60 -> 0.62Improved
add identity and perceptual losses (ablation)directions pipeline · Synthetic image pairs (5K), cross-subject (ablation)CSIM: 0.42 -> 0.52Improved
add eye and mouth losses (ablation)directions pipeline · Synthetic image pairs (5K), cross-subject (ablation)AED: 1.5 -> 1.2Improved
add real images to synthetic training (ablation)directions pipeline · VoxCeleb1 self-reenactment (training data ablation)CSIM: 0.60 -> 0.63Improved
fine-tune on paired video data (ablation)directions pipeline · VoxCeleb1 self-reenactment (training data ablation)ARD: 1.6 -> 1.1Improved
add style loss (ablation)LOR+ joint training · VoxCeleb1 self-reenactment, joint training (loss ablation)CSIM: 0.64 -> 0.66Improved
add cycle loss (ablation)LOR+ joint training · VoxCeleb1 self-reenactment, joint training (loss ablation)CSIM: 0.62 -> 0.66Improved
add single-attribute change in 50% of training samples (ablation)directions matrix · Disentanglement (synthetic, yaw changed)pitch error: 1.3 -> 1.2Improved
use LOR+ instead of LOR (joint training + FSR, no optimization)reenactment · VoxCeleb1 cross-subject reenactmentFID: 86.2 -> 78.4Improved

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