Research map / Vision-language and multimodal / Cross-modal learning
Cross-modal Latent Space Alignment for Image to Avatar Translation
Manuel Ladron de Guevara, Jose Echevarria, Yijun Li, Yannick Hold-Geoffroy, Cameron Smith, Daichi Ito
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use parametric cross-modal avatar generation instead of image-to-image GAN baselines | Avatar generation · Held-out avatars | L1: 0.2059 -> 0.1864 | Improved |
| use proposed method instead of U-GAT-IT in user study | Avatar generation · User study quality | preference %: 99.1% | Improved |
| add cross-modal alignment losses (ablation) | Mapping network F · Held-out avatars | LPIPS: 0.1876 -> 0.1835 | Improved |
| add weight alignment loss (ablation) | Mapping network F · Held-out avatars | LPIPS: 0.1852 -> 0.1835 | Improved |
| add MSE alignment term (ablation) | Mapping network F · Held-out avatars | L1: 0.1841 -> 0.1832 | No clear effect |
| add weight alignment loss (ablation) | Mapping network F · Held-out avatars | L1: 0.1833 -> 0.1832 | No clear effect |
| train mapping with fewer than 3000 paired samples (ablation) | Mapping network F · Qualitative (Figure 8) | identity and avatar quality | Hurt |
| train with 3995 pairs instead of 9970 (ablation) | Mapping network F · Qualitative (Figure 8) | identity preservation | No clear effect |
Closest papers by meaning
- PERSE: Personalized 3D Generative Avatars from A Single Portrait (CVPR 2025 · similarity 0.87)
- GeneAvatar: Generic Expression-Aware Volumetric Head Avatar Editing from a Single Image (CVPR 2024 · similarity 0.86)
- High-Fidelity 3D Head Avatars Reconstruction through Spatially-Varying Expression Conditioned Neural Radiance Field (AAAI 2024 · similarity 0.86)
- FRESA: Feedforward Reconstruction of Personalized Skinned Avatars from Few Images (CVPR 2025 · similarity 0.86)
- 3DToonify: Creating Your High-Fidelity 3D Stylized Avatar Easily from 2D Portrait Images (CVPR 2024 · similarity 0.86)
- NECA: Neural Customizable Human Avatar (CVPR 2024 · similarity 0.86)
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.
