Research map / Image generation / Editing and style transfer
GlassesGAN: Eyewear Personalization Using Synthetic Appearance Discovery and Targeted Subspace Modeling
Richard Plesh, Peter Peer, Vitomir Struc
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 GlassesGAN instead of the strongest baseline, GANSpace | Glasses editing · CelebA-HQ test | MSE: 0.0153 \pm 0.0064 -> 0.0029 \pm 0.0009 | Improved |
| Use GlassesGAN instead of the strongest baseline, GANSpace | Glasses editing · CelebA-HQ test | IDS: 0.4842 \pm 0.1060 -> 0.1707 \pm 0.0625 | Improved |
| Use GlassesGAN instead of the strongest baseline, GANSpace | Glasses editing · CelebA-HQ test | FID: 40.45 -> 26.02 | Improved |
| Use GlassesGAN instead of the strongest baseline, GANSpace | Glasses editing · SiblingsDB-HQf test | MSE: 0.0085 \pm 0.0030 -> 0.0029 \pm 0.0007 | Improved |
| Use GlassesGAN instead of the strongest baseline, GANSpace | Glasses editing · SiblingsDB-HQf test | IDS: 0.5047 \pm 0.0883 -> 0.1589 \pm 0.0478 | Improved |
| Use GlassesGAN instead of the strongest baseline, GANSpace | Glasses editing · SiblingsDB-HQf test | FID: 60.79 -> 45.83 | Improved |
| Add image blending with the original image (ablation) | GlassesGAN · Test images | IDS | Improved |
| Add image blending with the original image (ablation) | GlassesGAN · Test images | MSE: more than 5× reduction | Improved |
| Add image blending with the original image (ablation) | GlassesGAN · Test images | FID: more than 3× reduction | Improved |
| Add dynamic subspace initialization (ablation) | GlassesGAN · CelebA-HQ test | Edit Robustness Score: factor close to 5× reduction | Improved |
| Add Targeted Subspace Modeling instead of editing augmentation masks directly (ablation) | GlassesGAN · Glasses personalization | Edit flexibility | Improved |
| Add Targeted Subspace Modeling instead of editing augmentation masks directly (ablation) | GlassesGAN · Off-angle face edits | Edit robustness | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- InterfaceGAN — InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANs (TPAMI 2020) · better on 6 of 6 shared results
- MaskGAN — MaskGAN: Better Text Generation via Filling in the _______ (ICLR 2018) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Unwarping the Lens: A Physics-Grounded Approach to Video Glasses Removal (ECCV 2026 · similarity 0.86)
- Interpreting the Latent Space of GANs for Semantic Face Editing (CVPR 2020 · similarity 0.86)
- Face Generation and Editing With StyleGAN: A Survey (TPAMI 2024 · similarity 0.86)
- InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANs (TPAMI 2020 · similarity 0.86)
- SDGAN: Disentangling Semantic Manipulation for Facial Attribute Editing (AAAI 2024 · similarity 0.85)
- The Devil is in the Details: StyleFeatureEditor for Detail-Rich StyleGAN Inversion and High Quality Image Editing (CVPR 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.
