T2V-CompBench: A Comprehensive Benchmark for Compositional Text-to-video Generation

TL;DR

This paper introduces T2V-CompBench, a comprehensive benchmark for evaluating the ability of text-to-video models to generate videos with complex compositional attributes, highlighting current challenges and proposing new evaluation metrics.

Contribution

It presents the first dedicated benchmark for compositional text-to-video generation, including diverse evaluation metrics and a thorough analysis of existing models.

Findings

Current models struggle with compositional tasks

Proposed metrics correlate well with human judgment

Benchmark covers 7 compositional categories with 1400 prompts

Abstract

Text-to-video (T2V) generative models have advanced significantly, yet their ability to compose different objects, attributes, actions, and motions into a video remains unexplored. Previous text-to-video benchmarks also neglect this important ability for evaluation. In this work, we conduct the first systematic study on compositional text-to-video generation. We propose T2V-CompBench, the first benchmark tailored for compositional text-to-video generation. T2V-CompBench encompasses diverse aspects of compositionality, including consistent attribute binding, dynamic attribute binding, spatial relationships, motion binding, action binding, object interactions, and generative numeracy. We further carefully design evaluation metrics of multimodal large language model (MLLM)-based, detection-based, and tracking-based metrics, which can better reflect the compositional text-to-video generation quality of seven proposed categories with 1400 text prompts. The effectiveness of the proposed metrics is verified by correlation with human evaluations. We also benchmark various text-to-video generative models and conduct in-depth analysis across different models and various compositional categories. We find that compositional text-to-video generation is highly challenging for current models, and we hope our attempt could shed light on future research in this direction.