arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.01437cs.CVcs.AI
收藏

GPTSee:基于描述的相似性特征增强时刻检索与高光检测

GPTSee: Enhancing Moment Retrieval and Highlight Detection via Description-Based Similarity Features

School of Physics and Electronics, Hubei Normal University(湖北师范大学物理与电子科学学院);School of Electronic Science and Engineering, Nanjing University(南京大学电子科学与工程学院)

Yunzhuo Sun, Yifang Xu, Zien Xie, Yukun Shu, Sidan Du

AI总结:

针对现有时刻检索与高光检测方法未结合大语言模型的问题,提出两阶段模型GPTSee,利用MiniGPT-4生成视频描述并重写查询,通过语义相似度构建跨度锚点作为先验,实现更优的定位精度。

摘要:

时刻检索(MR)和高光检测(HD)旨在根据相应的自然语言查询,识别视频中的相关时刻和高光片段。大语言模型(LLM)已在多种计算机视觉任务中展现出优异性能,但现有MR和HD方法尚未与LLM相结合。本文提出一种新型两阶段模型,将LLM的输出作为第二阶段Transformer编码器-解码器的输入。首先,采用MiniGPT-4生成视频帧的详细描述并重写查询语句,作为新特征输入编码器;随后计算生成描述与重写查询之间的语义相似度;最后将连续的高相似度视频帧转换为跨度锚点,作为解码器的先验位置信息。实验表明,该方法取得了当前最优结果,仅使用跨度锚点和相似度分数作为输出时,定位精度优于Moment-DETR等传统方法。

英文摘要:

Moment retrieval (MR) and highlight detection (HD) aim to identify relevant moments and highlights in video from corresponding natural language query. Large language models (LLMs) have demonstrated proficiency in various computer vision tasks. However, existing methods for MR\&HD have not yet been integrated with LLMs. In this letter, we propose a novel two-stage model that takes the output of LLMs as the input to the second-stage transformer encoder-decoder. First, MiniGPT-4 is employed to generate the detailed description of the video frame and rewrite the query statement, fed into the encoder as new features. Then, semantic similarity is computed between the generated description and the rewritten queries. Finally, continuous high-similarity video frames are converted into span anchors, serving as prior position information for the decoder. Experiments demonstrate that our approach achieves a state-of-the-art result, and by using only span anchors and similarity scores as outputs, positioning accuracy outperforms traditional methods, like Moment-DETR.

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。

↑