T2I-CompBench++: An Enhanced and Comprehensive Benchmark for Compositional Text-to-image Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, Kaiyi, Duan, Chengqi, Sun, Kaiyue, Xie, Enze, Li, Zhenguo, Liu, Xihui
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913741233192960
author Huang, Kaiyi
Duan, Chengqi
Sun, Kaiyue
Xie, Enze
Li, Zhenguo
Liu, Xihui
author_facet Huang, Kaiyi
Duan, Chengqi
Sun, Kaiyue
Xie, Enze
Li, Zhenguo
Liu, Xihui
contents Despite the impressive advances in text-to-image models, they often struggle to effectively compose complex scenes with multiple objects, displaying various attributes and relationships. To address this challenge, we present T2I-CompBench++, an enhanced benchmark for compositional text-to-image generation. T2I-CompBench++ comprises 8,000 compositional text prompts categorized into four primary groups: attribute binding, object relationships, generative numeracy, and complex compositions. These are further divided into eight sub-categories, including newly introduced ones like 3D-spatial relationships and numeracy. In addition to the benchmark, we propose enhanced evaluation metrics designed to assess these diverse compositional challenges. These include a detection-based metric tailored for evaluating 3D-spatial relationships and numeracy, and an analysis leveraging Multimodal Large Language Models (MLLMs), i.e. GPT-4V, ShareGPT4v as evaluation metrics. Our experiments benchmark 11 text-to-image models, including state-of-the-art models, such as FLUX.1, SD3, DALLE-3, Pixart-$α$, and SD-XL on T2I-CompBench++. We also conduct comprehensive evaluations to validate the effectiveness of our metrics and explore the potential and limitations of MLLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2307_06350
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle T2I-CompBench++: An Enhanced and Comprehensive Benchmark for Compositional Text-to-image Generation
Huang, Kaiyi
Duan, Chengqi
Sun, Kaiyue
Xie, Enze
Li, Zhenguo
Liu, Xihui
Computer Vision and Pattern Recognition
Despite the impressive advances in text-to-image models, they often struggle to effectively compose complex scenes with multiple objects, displaying various attributes and relationships. To address this challenge, we present T2I-CompBench++, an enhanced benchmark for compositional text-to-image generation. T2I-CompBench++ comprises 8,000 compositional text prompts categorized into four primary groups: attribute binding, object relationships, generative numeracy, and complex compositions. These are further divided into eight sub-categories, including newly introduced ones like 3D-spatial relationships and numeracy. In addition to the benchmark, we propose enhanced evaluation metrics designed to assess these diverse compositional challenges. These include a detection-based metric tailored for evaluating 3D-spatial relationships and numeracy, and an analysis leveraging Multimodal Large Language Models (MLLMs), i.e. GPT-4V, ShareGPT4v as evaluation metrics. Our experiments benchmark 11 text-to-image models, including state-of-the-art models, such as FLUX.1, SD3, DALLE-3, Pixart-$α$, and SD-XL on T2I-CompBench++. We also conduct comprehensive evaluations to validate the effectiveness of our metrics and explore the potential and limitations of MLLMs.
title T2I-CompBench++: An Enhanced and Comprehensive Benchmark for Compositional Text-to-image Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2307.06350