MSG Score: Automated Video Verification for Reliable Multi-Scene Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yoon, Daewon, Lee, Hyeongseok, Shin, Wonsik, Han, Sangyu, Kwak, Nojun
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908946334220288
author Yoon, Daewon
Lee, Hyeongseok
Shin, Wonsik
Han, Sangyu
Kwak, Nojun
author_facet Yoon, Daewon
Lee, Hyeongseok
Shin, Wonsik
Han, Sangyu
Kwak, Nojun
contents While text-to-video diffusion models have advanced significantly, creating coherent long-form content remains unreliable due to stochastic sampling artifacts. This necessitates generating multiple candidates, yet verifying them creates a severe bottleneck; manual review is unscalable, and existing automated metrics lack the adaptability and speed required for runtime monitoring. Another critical issue is the trade-off between evaluation quality and run-time performance: metrics that best capture human-like judgment are often too slow to support iterative generation. These challenges, originating from the lack of an effective evaluation, motivate our work toward a novel solution. To address this, we propose a scalable automated verification framework for long-form video. First, we introduce the MSG(Multi-Scene Generation) score, a hierarchical attention-based metric that adaptively evaluates narrative and visual consistency. This serves as the core verifier within our CGS (Candidate Generation and Selection) framework, which automatically identifies and filters high-quality outputs. Furthermore, we introduce Implicit Insight Distillation (IID) to resolve the trade-off between evaluation reliability and inference speed, distilling complex metric insights into a lightweight student model. Our approach offers the first comprehensive solution for reliable and scalable long-form video production.
format Preprint
id arxiv_https___arxiv_org_abs_2411_19121
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MSG Score: Automated Video Verification for Reliable Multi-Scene Generation
Yoon, Daewon
Lee, Hyeongseok
Shin, Wonsik
Han, Sangyu
Kwak, Nojun
Computer Vision and Pattern Recognition
Artificial Intelligence
I.4
While text-to-video diffusion models have advanced significantly, creating coherent long-form content remains unreliable due to stochastic sampling artifacts. This necessitates generating multiple candidates, yet verifying them creates a severe bottleneck; manual review is unscalable, and existing automated metrics lack the adaptability and speed required for runtime monitoring. Another critical issue is the trade-off between evaluation quality and run-time performance: metrics that best capture human-like judgment are often too slow to support iterative generation. These challenges, originating from the lack of an effective evaluation, motivate our work toward a novel solution. To address this, we propose a scalable automated verification framework for long-form video. First, we introduce the MSG(Multi-Scene Generation) score, a hierarchical attention-based metric that adaptively evaluates narrative and visual consistency. This serves as the core verifier within our CGS (Candidate Generation and Selection) framework, which automatically identifies and filters high-quality outputs. Furthermore, we introduce Implicit Insight Distillation (IID) to resolve the trade-off between evaluation reliability and inference speed, distilling complex metric insights into a lightweight student model. Our approach offers the first comprehensive solution for reliable and scalable long-form video production.
title MSG Score: Automated Video Verification for Reliable Multi-Scene Generation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
I.4
url https://arxiv.org/abs/2411.19121