Research map / Image generation / Editing and style transfer
ByeGlassesGAN: Identity Preserving Eyeglasses Removal for Face Images
Yu-Hui Lee, Shang-Hong Lai
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 |
|---|---|---|---|
| add the segmentation decoder branch (ablation) | ByeGlassesGAN · MeGlass glasses removal | FID (lower better): 28.26 -> 27.14 | Improved |
| add the segmentation decoder branch (ablation) | ByeGlassesGAN · CelebA glasses removal | FID (lower better): 44.76 -> 42.97 | Improved |
| add identity extractor loss L_ID (ablation) | ByeGlassesGAN · MeGlass face recognition (Exp M) | TAR@FAR=1e-5: 0.5813 -> 0.6076 | Improved |
| add identity extractor loss L_ID (ablation) | ByeGlassesGAN · MeGlass face recognition (Exp M) | Rank-1: 0.8358 -> 0.8415 | Improved |
| add ByeGlassesGAN glasses removal before recognition | MobileFaceNet · MeGlass glasses probe | TAR@FAR=1e-5: 0.5708 -> 0.6076 | Improved |
| add ByeGlassesGAN glasses removal before recognition | MobileFaceNet · MeGlass glasses probe | Rank-1: 0.8275 -> 0.8415 | Improved |
| add CycleGAN glasses removal before recognition | MobileFaceNet · MeGlass glasses probe | TAR@FAR=1e-5: 0.5708 -> 0.5329 | Hurt |
| add ELEGANT glasses removal before recognition | MobileFaceNet · MeGlass glasses probe | TAR@FAR=1e-5: 0.5708 -> 0.3719 | Hurt |
| add pix2pix glasses removal before recognition | MobileFaceNet · MeGlass glasses probe | TAR@FAR=1e-5: 0.5708 -> 0.5251 | Hurt |
| add SaGAN glasses removal before recognition | MobileFaceNet · MeGlass glasses probe | TAR@FAR=1e-5: 0.5708 -> 0.5684 | Hurt |
| add ByeGlassesGAN removal before recognition | Identity Extractor (Exp R) · MeGlass glasses probe | TAR@FAR=1e-5: 0.6292 -> 0.6819 | Improved |
| add ByeGlassesGAN removal to all images in mixed gallery and probe | MobileFaceNet · MeGlass mixed gallery/probe | TAR@FAR=1e-5: 0.5927 -> 0.6539 | 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.
- ELEGANT — ELEGANT: Exchanging Latent Encodings with GAN for Transferring Multiple Face Attributes (ECCV 2018) · better on 5 of 5 shared results
- StarGAN — StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation (CVPR 2018) · better on 2 of 2 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)
- Portrait Eyeglasses and Shadow Removal by Leveraging 3D Synthetic Data (CVPR 2022 · similarity 0.86)
- Face Image Reflection Removal (IJCV 2020 · similarity 0.85)
- Eyeglasses 3D shape reconstruction from a single face image (ECCV 2020 · similarity 0.84)
- Anti-Glare: Tightly Constrained Optimization for Eyeglass Reflection Removal (CVPR 2017 · similarity 0.84)
- SmartEraser: Remove Anything from Images using Masked-Region Guidance (CVPR 2025 · similarity 0.81)
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.
