OpenS2V-Nexus: A Detailed Benchmark and Million-Scale Dataset for Subject-to-Video Generation

TL;DR

OpenS2V-Nexus provides a comprehensive benchmark and a large-scale dataset for subject-to-video generation, enabling more accurate assessment and development of models that produce subject-consistent videos.

Contribution

The paper introduces OpenS2V-Nexus, including a fine-grained benchmark and a million-scale dataset, advancing evaluation and data resources for subject-to-video generation.

Findings

Evaluated 18 S2V models highlighting their strengths and weaknesses.

Proposed three automatic metrics for assessing generated videos.

Created the first large-scale open-source S2V dataset with 5 million samples.

Abstract

Subject-to-Video (S2V) generation aims to create videos that faithfully incorporate reference content, providing enhanced flexibility in the production of videos. To establish the infrastructure for S2V generation, we propose OpenS2V-Nexus, consisting of (i) OpenS2V-Eval, a fine-grained benchmark, and (ii) OpenS2V-5M, a million-scale dataset. In contrast to existing S2V benchmarks inherited from VBench that focus on global and coarse-grained assessment of generated videos, OpenS2V-Eval focuses on the model's ability to generate subject-consistent videos with natural subject appearance and identity fidelity. For these purposes, OpenS2V-Eval introduces 180 prompts from seven major categories of S2V, which incorporate both real and synthetic test data. Furthermore, to accurately align human preferences with S2V benchmarks, we propose three automatic metrics, NexusScore, NaturalScore and GmeScore, to separately quantify subject consistency, naturalness, and text relevance in generated videos. Building on this, we conduct a comprehensive evaluation of 18 representative S2V models, highlighting their strengths and weaknesses across different content. Moreover, we create the first open-source large-scale S2V generation dataset OpenS2V-5M, which consists of five million high-quality 720P subject-text-video triples. Specifically, we ensure subject-information diversity in our dataset by (1) segmenting subjects and building pairing information via cross-video associations and (2) prompting GPT-Image-1 on raw frames to synthesize multi-view representations. Through OpenS2V-Nexus, we deliver a robust infrastructure to accelerate future S2V generation research.