DreamBench++: A Human-Aligned Benchmark for Personalized Image Generation

TL;DR

DreamBench++ is a new benchmark for personalized image generation that uses advanced GPT models to provide human-aligned, automated evaluation, addressing the limitations of existing automated and human-based assessments.

Contribution

It introduces a systematic prompt design for GPT models to enable human-aligned and self-aligned evaluation of personalized image generation models.

Findings

GPT-based evaluation aligns better with human judgment.

Benchmarking reveals strengths and weaknesses of current models.

The dataset enhances evaluation diversity and robustness.

Abstract

Personalized image generation holds great promise in assisting humans in everyday work and life due to its impressive ability to creatively generate personalized content across various contexts. However, current evaluations either are automated but misalign with humans or require human evaluations that are time-consuming and expensive. In this work, we present DreamBench++, a human-aligned benchmark that advanced multimodal GPT models automate. Specifically, we systematically design the prompts to let GPT be both human-aligned and self-aligned, empowered with task reinforcement. Further, we construct a comprehensive dataset comprising diverse images and prompts. By benchmarking 7 modern generative models, we demonstrate that DreamBench++ results in significantly more human-aligned evaluation, helping boost the community with innovative findings.