Research map / Image generation / Editing and style transfer
ManiTrans: Entity-Level Text-Guided Image Manipulation via Token-Wise Semantic Alignment and Generation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 ManiTrans instead of ManiGAN | text-guided entity manipulation · CUB | R@10: 10.49 -> 34.82 | Helped |
| Use ManiTrans instead of ManiGAN | text-guided entity manipulation · Oxford | R@10: 14.21 -> 36.49 | Helped |
| Use ManiTrans instead of ManiGAN | text-guided entity manipulation · COCO | R@10: 14.50 -> 21.32 | Helped |
| Use ManiTrans instead of Lightweight-GAN | text-guided entity manipulation · COCO | IS: 24.80 0.94 -> 21.45 0.41 | Hurt |
| Add grayscale vision guidance (ablation) | ManiTrans · CUB | CLIP-score: 24.02 -> 23.56 | Hurt |
| Add grayscale vision guidance (ablation) | ManiTrans · CUB | L2-error: 0.02 -> 0.01 | Helped |
| Add CLIP semantic loss (ablation) | ManiTrans · CUB | CLIP-score: 22.19 -> 23.56 | Helped |
| Add CLIP semantic loss (ablation) | ManiTrans · CUB | R@10: 16.06 -> 34.82 | Helped |
| Use entity-level semantic alignment instead of word-patch alignment (ablation) | ManiTrans inference · CUB | R@10: 32.89 -> 34.82 | Helped |
| Use entity-level semantic alignment instead of word-patch alignment (ablation) | ManiTrans inference · CUB | CLIP-score: 23.53 -> 23.56 | No clear effect |
Closest papers by meaning
- Progressive Text-to-Image Diffusion with Soft Latent Direction (AAAI 2024 · similarity 0.89)
- Prompt Augmentation for Self-supervised Text-guided Image Manipulation (CVPR 2024 · similarity 0.88)
- Text-Driven Image Editing via Learnable Regions (CVPR 2024 · similarity 0.87)
- Token Merging for Training-Free Semantic Binding in Text-to-Image Synthesis (NeurIPS 2024 · similarity 0.87)
- Multi-Region Text-Driven Manipulation of Diffusion Imagery (AAAI 2024 · similarity 0.87)
- Learning to Manipulate Artistic Images (AAAI 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.