Research map / Image generation / Editing and style transfer
End-to-End Visual Editing with a Generatively Pre-trained Artist
Andrew Brown, Cheng-Yang Fu, Omkar Parkhi, Tamara L. Berg, Andrea Vedaldi
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 pos-augment (ablation) | E2EVE · Dresses-7m (block-edits) | FID (Image): 15.605 -> 14.951 | Improved |
| add size-augment (ablation) | E2EVE · Dresses-7m (block-edits) | FID (Edit-R): 26.513 -> 24.743 | Improved |
| add size-augment (ablation) | E2EVE · Dresses-7m (block-edits) | Recall@1: 0.887 -> 0.797 | Hurt |
| lower alpha from 0.8 to 0.4 (ablation) | E2EVE · Dresses-7m (block-edits) | FID (Image): 17.241 -> 13.967 | Improved |
| lower alpha from 0.6 to 0.4 (ablation) | E2EVE · Dresses-7m (block-edits) | Recall@1: 0.920 -> 0.811 | Hurt |
| set alpha to 0.0 (ablation) | E2EVE · Dresses-7m (block-edits) | Recall@1: 0.811 -> 0.071 | Hurt |
| vary alpha (ablation) | E2EVE · Dresses-7m (block-edits) | L1 locality: 0.056 | No clear effect |
| add similarity filtering of samples (ablation) | E2EVE · Dresses-7m (block-edits) | Recall@1: 0.611 -> 0.797 | Improved |
| add similarity filtering of samples (ablation) | EdiBERT · LSUN Bedrooms/FFHQ | Recall@1: 0.356 -> 0.537 | Improved |
| use two VQ-GANs instead of one (ablation) | E2EVE · Dresses-7m (block-edits) | Recall@1: 0.720 -> 0.797 | Improved |
| use two VQ-GANs instead of one (ablation) | E2EVE · Dresses-7m (block-edits) | FID (Edit-R): 23.916 -> 24.743 | Hurt |
| add pos and size augmentation (ablation) | E2EVE · Dresses-7m (block-edits) | NLL: 1.704 -> 1.448 | Improved |
Closest papers by meaning
- Learning to Edit Visual Programs with Self-Supervision (NeurIPS 2024 · similarity 0.88)
- Zero-shot Image Editing with Reference Imitation (NeurIPS 2024 · similarity 0.88)
- EditAR: Unified Conditional Generation with Autoregressive Models (CVPR 2025 · similarity 0.87)
- UltraEdit: Instruction-based Fine-Grained Image Editing at Scale (NeurIPS 2024 · similarity 0.87)
- Prompt Augmentation for Self-supervised Text-guided Image Manipulation (CVPR 2024 · similarity 0.87)
- Learning Where to Edit Vision Transformers (NeurIPS 2024 · similarity 0.87)
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.
