Research map / Image generation / Text-to-image diffusion
PixTalk: Controlling Photorealistic Image Processing and Editing with Language
Marcos V. Conde, Zihao Lu, Radu Timofte
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 task classification loss on the text embedding (ablation) | PixTalk text encoder · PixTalk training | training stability | Improved |
| select only top-10 color-correction kernels at inference | PixTalk color enhancement block · PixTalk | PSNR | No clear effect |
| use sigmoid controller instead of softmax mixture of experts (ablation) | PixTalk language controller · PixTalk | block selection | Improved |
| replace the learned text controller with LLaVA-1.5 on a frozen PixTalk | PixTalk · PixTalk Benchmark | PSNR: 37.80 -> 37.90 | No clear effect |
| use PixTalk instead of InstructIR | text-guided image processing · PixTalk Benchmark | PSNR: 38.10 -> 37.80 | Hurt |
| use PixTalk instead of zero-shot InstructPix2Pix | text-guided image editing · PixTalk Benchmark | PSNR: 22.68 -> 37.80 | Improved |
| use PixTalk instead of zero-shot Ledits++ | text-guided image editing · PixTalk Benchmark | PSNR: 24.68 -> 37.80 | Improved |
| use PixTalk instead of CNILUT | text-guided image processing · PixTalk Benchmark | SSIM: 0.960 -> 0.950 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- InstructIR — InstructIR: High-Quality Image Restoration Following Human Instructions (ECCV 2024) · better on 1 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
- PixWizard: Versatile Image-to-Image Visual Assistant with Open-Language Instructions (ICLR 2025 · similarity 0.86)
- PIXELS: Progressive Image Xemplar-based Editing with Latent Surgery (AAAI 2025 · similarity 0.85)
- InsightEdit: Towards Better Instruction Following for Image Editing (CVPR 2025 · similarity 0.85)
- SmartEdit: Exploring Complex Instruction-based Image Editing with Multimodal Large Language Models (CVPR 2024 · similarity 0.85)
- ImgEdit: A Unified Image Editing Dataset and Benchmark (NeurIPS 2025 · similarity 0.84)
- PixPerfect: Seamless Latent Diffusion Local Editing with Discriminative Pixel-Space Refinement (NeurIPS 2025 · 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.
