Research map / 3D vision / Face and shape
A Multi-Implicit Neural Representation for Fonts
Pradyumna Reddy, Zhifei Zhang, Zhaowen Wang, Matthew Fisher, Hailin Jin, Niloy Mitra
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 multi-implicit rendering instead of Im2Vec vector output | font reconstruction model · Reconstruction @1024px | s-IoU: .7294 -> .9049 | Improved |
| use multi-implicit model instead of ImageVAE | font reconstruction model · Reconstruction @1024px | s-IoU: .8494 -> .9049 | Improved |
| use multi-implicit model instead of ImageVAE | font reconstruction model · Reconstruction @1024px | MSE: .0186 -> .0218 | Hurt |
| use multi-implicit model instead of Im2Vec | font reconstruction model · Reconstruction @1024px | MSE: .0571 -> .0218 | Improved |
| use multi-implicit model instead of ImageVAE | font interpolation model · Interpolation @1024px | s-IoU: .7734 -> .8234 | Improved |
| use multi-implicit model instead of ImageVAE | font interpolation model · Interpolation @1024px | MSE: .0185 -> .0343 | Hurt |
| use multi-implicit model instead of Im2Vec | font interpolation model · Interpolation @1024px | MSE: .0473 -> .0343 | Improved |
| use multi-implicit model instead of Attr2Font | font generation model · Font generation @1024px | s-IoU: .6560 -> .8471 | Improved |
| use multi-implicit model instead of Attr2Font | font generation model · Font generation @1024px | MSE: .2563 -> .1083 | Improved |
| use multi-implicit model instead of DeepSVG | font generation model · Font generation @128px | s-IoU: .3584 -> .8429 | Improved |
| increase test resolution from 128 to 1024 for ImageVAE (bilinear upsampling) | ImageVAE · Reconstruction | MSE: .0072 -> .0186 | Hurt |
| increase test resolution from 128 to 1024 for multi-implicit model | multi-implicit model · Reconstruction | s-IoU: .8750 -> .9049 | Improved |
Closest papers by meaning
- Joint Implicit Neural Representation for High-fidelity and Compact Vector Fonts (ICCV 2023 · similarity 0.89)
- Few-Shot Font Generation With Weakly Supervised Localized Representations (TPAMI 2022 · similarity 0.87)
- VecFusion: Vector Font Generation with Diffusion (CVPR 2024 · similarity 0.86)
- FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learning (AAAI 2024 · similarity 0.85)
- Generate Like Experts: Multi-Stage Font Generation by Incorporating Font Transfer Process into Diffusion Models (CVPR 2024 · similarity 0.85)
- IF-Font: Ideographic Description Sequence-Following Font Generation (NeurIPS 2024 · similarity 0.84)
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.
