Research map / Image generation / Text-to-image diffusion
RAVE: Residual Vector Embedding for CLIP-Guided Backlit Image Enhancement
Tatiana Gaintseva, Martin Benning, Gregory Slabaugh
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 |
|---|---|---|---|
| learn guidance vectors directly in CLIP latent space instead of text prompts (CLIP-LIT-Latent) | CLIP-LIT · BAID test (paired) | PSNR: 21.93 -> 21.84 | No clear effect |
| replace prompt tuning with fixed residual vector (RAVE) | CLIP-LIT · BAID test (paired) | FID: 41.16 -> 36.01 | Improved |
| replace prompt tuning with residual vector (RAVE) on unpaired data | CLIP-LIT · BAID test (unpaired) | PSNR: 21.59 -> 20.39 | Hurt |
| replace prompt tuning with residual vector (RAVE) on unpaired data | CLIP-LIT · BAID test (unpaired) | FID: 46.49 -> 40.11 | Improved |
| remove dataset-bias token semantics from residual vector (RAVE shifted) (ablation) | RAVE (unpaired) · BAID test (unpaired) | PSNR: 20.39 -> 21.37 | Improved |
| use CLIP text embeddings of dark-related tokens as guidance instead of data-derived residual vector (ablation) | RAVE · BAID | enhancement quality (visual) | Hurt |
| use RAVE single-stage training instead of iterative prompt refinement | CLIP-LIT · training time | convergence speed: up to 25 times faster | Improved |
| use RAVE vs Diff-Retinex on paired data | backlit enhancement · BAID test (paired) | PSNR: 22.07 -> 22.26 | 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.
- Diff-Retinex — Diff-Retinex: Rethinking Low-light Image Enhancement with A Generative Diffusion Model (ICCV 2023) · better on 2 of 2 shared results
- Retinexformer — Retinexformer: One-stage Retinex-based Transformer for Low-light Image Enhancement (ICCV 2023) · better on 1 of 1 shared result
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
- Instruct-CLIP: Improving Instruction-Guided Image Editing with Automated Data Refinement Using Contrastive Learning (CVPR 2025 · similarity 0.85)
- MLIP: Efficient Multi-Perspective Language-Image Pretraining with Exhaustive Data Utilization (ICML 2024 · similarity 0.85)
- LP++: A Surprisingly Strong Linear Probe for Few-Shot CLIP (CVPR 2024 · similarity 0.85)
- un$^2$CLIP: Improving CLIP's Visual Detail Capturing Ability via Inverting unCLIP (NeurIPS 2025 · similarity 0.85)
- Controlling Vision-Language Models for Multi-Task Image Restoration (ICLR 2024 · similarity 0.85)
- Demystifying CLIP Data (ICLR 2024 · similarity 0.85)
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.
