arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2108.09635cs.CV
收藏

StarVQA:用于视频质量评估的时空注意力

StarVQA: Space-Time Attention for Video Quality Assessment

Guangzhou University(广州大学);Guangdong University of Education(广东第二师范学院);Xidian University(西安电子科技大学);Harbin Institute of Technology(哈尔滨工业大学)

Fengchuang Xing, Yuan-Gen Wang, Hanpin Wang, Leida Li, Guopu Zhu

摘要:

注意力机制在计算机视觉中得到了广泛的应用。然而,其应用于视频质量评估(VQA)还没有报道。由于原始参考和拍摄失真的未知性,评估野外视频的质量具有挑战性。本文提出了一种新的\underline{s}空间- \underline{t}时间\underline{a}约束网络,用于\underline{r}的\underline{VQA}问题,称为StarVQA. StarVQA通过交替连接划分的时空注意力来构建一个Transformer。为了使Transformer架构适应训练,StarVQA通过将平均意见得分(MOS)编码为概率向量并嵌入特殊的向量化标签令牌作为可学习变量来设计向量化回归损失。为了捕获视频序列的长距离时空依赖性,StarVQA将每个补丁的时空位置信息编码到Transformer的输入。在实际的野外视频数据集上进行了各种实验,包括LIVE-VQC,KoNViD-1 k,LSVQ和LSVQ-1080 p。实验结果表明,所提出的StarVQA的国家的最先进的优越性。代码和模型将在:https://github.com/DVL/StarVQA。

英文摘要:

The attention mechanism is blooming in computer vision nowadays. However, its application to video quality assessment (VQA) has not been reported. Evaluating the quality of in-the-wild videos is challenging due to the unknown of pristine reference and shooting distortion. This paper presents a novel \underline{s}pace-\underline{t}ime \underline{a}ttention network fo\underline{r} the \underline{VQA} problem, named StarVQA. StarVQA builds a Transformer by alternately concatenating the divided space-time attention. To adapt the Transformer architecture for training, StarVQA designs a vectorized regression loss by encoding the mean opinion score (MOS) to the probability vector and embedding a special vectorized label token as the learnable variable. To capture the long-range spatiotemporal dependencies of a video sequence, StarVQA encodes the space-time position information of each patch to the input of the Transformer. Various experiments are conducted on the de-facto in-the-wild video datasets, including LIVE-VQC, KoNViD-1k, LSVQ, and LSVQ-1080p. Experimental results demonstrate the superiority of the proposed StarVQA over the state-of-the-art. Code and model will be available at: https://github.com/DVL/StarVQA.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑