PersonaShot: Benchmarking Person-Centric Narrative Continuity in Multi-Shot Video Generation

2026-08-17

AI Summary

This study addresses the absence of evaluation benchmarks for human narrative coherence in multi-shot video generation by constructing the first human-centric dataset comprising over one thousand clips and sixteen metrics. Through multimodal knowledge distillation and hierarchical temporal modeling, we propose a lightweight specialized evaluator aligned with human experts to precisely quantify cross-shot consistency in physical and emotional states. Our analysis reveals significant deficiencies in state-of-the-art models regarding narrative continuity. Demonstrating high correlation with expert judgment, this evaluator provides a systematic profiling of current limitations and establishes clear directions for enhancing the narrative capabilities of video generation systems.

Abstract

Video generation is rapidly evolving from single-shot clips to multi-shot narratives, where the human character serves as the core narrative anchor. However, existing benchmarks mainly assess character appearance or individual-shot quality, without measuring whether physical and emotional states remain coherent across cuts. They also rarely provide criterion-specific evaluation methods, although physical continuity, facial dynamics, and cinematic relations require different visual, temporal, and relational evidence. To address these limitations, we introduce PersonaShot, the first person-centric benchmark for narrative continuity in multi-shot video generation. PersonaShot contains approximately 1,000 multi-shot segments and 16 metrics spanning physical continuity, affective dynamics, and cinematic grammar. \textbf{\textit{1)} Narrative Continuity Benchmark:} We evaluate character coherence across three temporal levels: within-shot states, cross-shot transitions, and sequence-level trajectories. \textbf{\textit{2)} Human-Aligned Specialist Evaluators:} We distill reasoning from a large multimodal teacher into lightweight criterion-specific evaluators, each grounded in the visual, temporal, or relational evidence required by its metric, and align them with expert human judgments. \textbf{\textit{3)} Systematic Evaluation and Insights:} Our evaluation reveals distinct capability profiles across state-of-the-art models and a clear gap between perceptual quality and cross-shot narrative continuity. Even visually compelling videos frequently exhibit physical-state resets, abrupt affective shifts, and broken cinematic relations across shots. Human studies further demonstrate strong agreement between our evaluators and expert judgments.

Problem

• Multi-Shot Video Generation
• Narrative Continuity
• Person-Centric Evaluation
• Benchmark

Innovation

• Narrative Continuity Benchmark
• Person-Centric Evaluation
• Human-Aligned Specialist Evaluators
• Multi-Shot Video Generation
• Knowledge Distillation