Advancing Video Quality Assessment for AIGC

TL;DR

This paper introduces a new video quality assessment method tailored for AI-generated videos, combining a novel loss function and S2CNet to better evaluate and improve frame consistency and content retention.

Contribution

The paper presents a novel loss function and S2CNet technique specifically designed for assessing and enhancing AI-generated video quality, addressing gaps in existing evaluation methods.

Findings

Outperforms existing VQA methods on AIGC Video dataset

Achieves 3.1% higher PLCC than previous state-of-the-art

Effectively reduces inter-frame quality discrepancies

Abstract

In recent years, AI generative models have made remarkable progress across various domains, including text generation, image generation, and video generation. However, assessing the quality of text-to-video generation is still in its infancy, and existing evaluation frameworks fall short when compared to those for natural videos. Current video quality assessment (VQA) methods primarily focus on evaluating the overall quality of natural videos and fail to adequately account for the substantial quality discrepancies between frames in generated videos. To address this issue, we propose a novel loss function that combines mean absolute error with cross-entropy loss to mitigate inter-frame quality inconsistencies. Additionally, we introduce the innovative S2CNet technique to retain critical content, while leveraging adversarial training to enhance the model's generalization capabilities. Experimental results demonstrate that our method outperforms existing VQA techniques on the AIGC Video dataset, surpassing the previous state-of-the-art by 3.1% in terms of PLCC.