arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2202.09277cs.CVcs.AIcs.LG
收藏

(2.5用于视频问答的+1)D时空场景图

(2.5+1)D Spatio-Temporal Scene Graphs for Video Question Answering

Mitsubishi Electric Research Labs (MERL)(三菱电机研究实验室)

Anoop Cherian, Chiori Hori, Tim K. Marks, Jonathan Le Roux

摘要:

基于视频的推理任务,如视频问答(QA)的时空场景图方法,通常为每个视频帧构建这样的图。这些方法通常忽略了以下事实:视频本质上是在3D空间中发生的事件的2D“视图”的序列,并且3D场景的语义因此可以从帧到帧进行。利用这一洞察力,我们提出了一个(2.5+1)D场景图表示,以更好地捕捉视频内部的时空信息流。具体来说,我们首先创建一个2.5D通过使用现成的2D到3D变换模块将每个2D帧变换为具有推断的3D结构来生成(伪3D)场景图,之后,我们将视频帧注册到共享的(2.5+1)D时空空间,并将其内的每个2D场景图接地。(2.5+1)维图根据其内部对象是否经常在世界中移动而被划分为静态子图和动态子图。动态图中的节点被丰富有捕捉它们与其他图节点的交互的运动特征。接下来,对于视频QA任务,我们提出了一种新的基于transformer的推理流水线,该流水线将(2.5+1)D图嵌入到时空分层潜在空间中,其中子图及其交互以不同的粒度捕获。为了证明我们的方法的有效性,我们在NExT-QA和AVSD-QA数据集上进行了实验。我们的研究结果表明,我们提出的(2.5+1)D表示导致更快的训练和推理,而我们的分层模型在视频QA任务上表现出优于现有技术的性能。

英文摘要:

Spatio-temporal scene-graph approaches to video-based reasoning tasks, such as video question-answering (QA), typically construct such graphs for every video frame. These approaches often ignore the fact that videos are essentially sequences of 2D "views" of events happening in a 3D space, and that the semantics of the 3D scene can thus be carried over from frame to frame. Leveraging this insight, we propose a (2.5+1)D scene graph representation to better capture the spatio-temporal information flows inside the videos. Specifically, we first create a 2.5D (pseudo-3D) scene graph by transforming every 2D frame to have an inferred 3D structure using an off-the-shelf 2D-to-3D transformation module, following which we register the video frames into a shared (2.5+1)D spatio-temporal space and ground each 2D scene graph within it. Such a (2.5+1)D graph is then segregated into a static sub-graph and a dynamic sub-graph, corresponding to whether the objects within them usually move in the world. The nodes in the dynamic graph are enriched with motion features capturing their interactions with other graph nodes. Next, for the video QA task, we present a novel transformer-based reasoning pipeline that embeds the (2.5+1)D graph into a spatio-temporal hierarchical latent space, where the sub-graphs and their interactions are captured at varied granularity. To demonstrate the effectiveness of our approach, we present experiments on the NExT-QA and AVSD-QA datasets. Our results show that our proposed (2.5+1)D representation leads to faster training and inference, while our hierarchical model showcases superior performance on the video QA task versus the state of the art.

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。

↑