RealGen: Photorealistic Text-to-Image Generation via Detector-Guided Rewards

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ye, Junyan, Zhu, Leiqi, Guo, Yuncheng, Jiang, Dongzhi, Huang, Zilong, Zhang, Yifan, Yan, Zhiyuan, Fu, Haohuan, He, Conghui, Li, Weijia
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917113926516736
author Ye, Junyan
Zhu, Leiqi
Guo, Yuncheng
Jiang, Dongzhi
Huang, Zilong
Zhang, Yifan
Yan, Zhiyuan
Fu, Haohuan
He, Conghui
Li, Weijia
author_facet Ye, Junyan
Zhu, Leiqi
Guo, Yuncheng
Jiang, Dongzhi
Huang, Zilong
Zhang, Yifan
Yan, Zhiyuan
Fu, Haohuan
He, Conghui
Li, Weijia
contents With the continuous advancement of image generation technology, advanced models such as GPT-Image-1 and Qwen-Image have achieved remarkable text-to-image consistency and world knowledge However, these models still fall short in photorealistic image generation. Even on simple T2I tasks, they tend to produce " fake" images with distinct AI artifacts, often characterized by "overly smooth skin" and "oily facial sheens". To recapture the original goal of "indistinguishable-from-reality" generation, we propose RealGen, a photorealistic text-to-image framework. RealGen integrates an LLM component for prompt optimization and a diffusion model for realistic image generation. Inspired by adversarial generation, RealGen introduces a "Detector Reward" mechanism, which quantifies artifacts and assesses realism using both semantic-level and feature-level synthetic image detectors. We leverage this reward signal with the GRPO algorithm to optimize the entire generation pipeline, significantly enhancing image realism and detail. Furthermore, we propose RealBench, an automated evaluation benchmark employing Detector-Scoring and Arena-Scoring. It enables human-free photorealism assessment, yielding results that are more accurate and aligned with real user experience. Experiments demonstrate that RealGen significantly outperforms general models like GPT-Image-1 and Qwen-Image, as well as specialized photorealistic models like FLUX-Krea, in terms of realism, detail, and aesthetics. The code is available at https://github.com/yejy53/RealGen.
format Preprint
id arxiv_https___arxiv_org_abs_2512_00473
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RealGen: Photorealistic Text-to-Image Generation via Detector-Guided Rewards
Ye, Junyan
Zhu, Leiqi
Guo, Yuncheng
Jiang, Dongzhi
Huang, Zilong
Zhang, Yifan
Yan, Zhiyuan
Fu, Haohuan
He, Conghui
Li, Weijia
Computer Vision and Pattern Recognition
Artificial Intelligence
With the continuous advancement of image generation technology, advanced models such as GPT-Image-1 and Qwen-Image have achieved remarkable text-to-image consistency and world knowledge However, these models still fall short in photorealistic image generation. Even on simple T2I tasks, they tend to produce " fake" images with distinct AI artifacts, often characterized by "overly smooth skin" and "oily facial sheens". To recapture the original goal of "indistinguishable-from-reality" generation, we propose RealGen, a photorealistic text-to-image framework. RealGen integrates an LLM component for prompt optimization and a diffusion model for realistic image generation. Inspired by adversarial generation, RealGen introduces a "Detector Reward" mechanism, which quantifies artifacts and assesses realism using both semantic-level and feature-level synthetic image detectors. We leverage this reward signal with the GRPO algorithm to optimize the entire generation pipeline, significantly enhancing image realism and detail. Furthermore, we propose RealBench, an automated evaluation benchmark employing Detector-Scoring and Arena-Scoring. It enables human-free photorealism assessment, yielding results that are more accurate and aligned with real user experience. Experiments demonstrate that RealGen significantly outperforms general models like GPT-Image-1 and Qwen-Image, as well as specialized photorealistic models like FLUX-Krea, in terms of realism, detail, and aesthetics. The code is available at https://github.com/yejy53/RealGen.
title RealGen: Photorealistic Text-to-Image Generation via Detector-Guided Rewards
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2512.00473