PixTalk: Controlling Photorealistic Image Processing and Editing with Language
Marcos V. Conde, Zihao Lu, Radu Timofte
2025年份
4被引次数
1顶会引用
摘要
The enitre image looks flat, I want an image with higher contrast.
The image looks too dark and cold, make it a bit warmer and brighter.
Can you make my photo look like in winter?
Figure 1. Given an image and text prompts for how to edit the image, our multi-task image processing model edits the image considering the human instructions. PixTalk, covers over 40 transformations that can be commonly found in photography software (e.g. Adobe Lightroom), and it generalizes of-the-shelf on real-world photographs from diverse cameras and sensors.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper1
问问它们各自怎么用它它引用的顶会 Paper24
- LoRA: Low-Rank Adaptation of Large Language ModelsEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu 等ICLR 2022 · 被引用 18,833 次
- Visual Instruction TuningHaotian Liu, Chunyuan Li, Qingyang Wu, Yong Jae LeeNeurIPS 2023 · 被引用 11,349 次
- SDEdit: Guided Image Synthesis and Editing with Stochastic Differential EquationsChenlin Meng, Yutong He, Yang Song, Jiaming Song 等ICLR 2022 · 被引用 2,128 次
- Prompt-to-Prompt Image Editing with Cross-Attention ControlAmir Hertz, Ron Mokady, Jay Tenenbaum, Kfir Aberman 等ICLR 2023 · 被引用 361 次
- All-In-One Image Restoration for Unknown CorruptionBoyun Li, Xiao Liu, Peng Hu, Zhongqin Wu 等CVPR 2022 · 被引用 338 次
相关 Paper
- Diff-Plugin: Revitalizing Details for Diffusion-Based Low-Level TasksYuhao Liu, Zhanghan Ke, Fang Liu, Nanxuan Zhao 等CVPR 2024
- InstructPix2Pix: Learning to Follow Image Editing InstructionsTim Brooks, Aleksander Holynski, Alexei A. EfrosCVPR 2023
- Instruction-based Image Manipulation by Watching How Things MoveMingdeng Cao, Xuaner Zhang, Yinqiang Zheng, Zhihao XiaCVPR 2025
- Prompt-Guided Image-Adaptive Neural Implicit Lookup Tables for Interpretable Image EnhancementSatoshi KosugiACM MM 2024 · 被引用 2 次
- Unifying Image Processing as Visual Prompting Question AnsweringYihao Liu, Xiangyu Chen, Xianzheng Ma, Xintao Wang 等ICML 2024 · 被引用 37 次
