Viewpoint-Aware Video Summarization

By Atsushi Kanehira, Luc Van Gool, Yoshitaka Ushiku, Tatsuya Harada. 2018

39 citations recorded by the source corpus.

This paper introduces a novel variant of video summarization, namely building a summary that depends on the particular aspect of a video the viewer focuses on. We refer to this as viewpoint. To infer what the desired viewpoint may be, we assume that several other videos are available, especially groups of videos, e.g., as folders on a person's phone or laptop. The semantic similarity between videos in a group vs. the dissimilarity between groups is used to produce viewpoint-specific summaries. For considering similarity as well as avoiding redundancy, output summary should be (A) diverse, (B) representative of videos in the same group, and (C) discriminative against videos in the different groups. To satisfy these requirements (A)-(C) simultaneously, we proposed a novel video summarization method from multiple groups of videos. Inspired by Fisher's discriminant criteria, it selects summary by optimizing the combination of three terms (a) inner-summary, (b) inner-group, and (c) between-group variances defined on the feature representation of summary, which can simply represent (A)-(C). Moreover, we developed a novel dataset to investigate how well the generated summary reflects the underlying viewpoint. Quantitative and qualitative experiments conducted on the dataset demonstrate the effectiveness of proposed method.

Subfields: Computer Vision and Pattern Recognition, Sociology and Political Science, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Sociology and Political Science

In Signal Processing

Citations in the corpus

Learning to Explain With Complemental Examples (2018, background): “Prototype selection [25, 7, 13, 16, 25, 10] or machine teaching [18] can be considered as example-based explanations.”
A Practical Online Framework for Extracting Running Video Summaries under a Fixed Memory Budget (2021, background): “Kanehira et al. [17] explore video summarization in the context of view-point extracted from multiple videos.”
Multimodal Explanations by Predicting Counterfactuality in Videos (2018, methodology): “Beside works of attempting counterfactual explanations by example selection [10, 31, 18], [2] conducted research similar to the present study, stating an application for grounding visual explanation to counterfactual explanation, where the textual explanations (without the region) are generated by c”
CLIP-It! Language-Guided Video Summarization (2021, background): “Other works [13, 40] similarly condition the summary generation on keyword based queries but haven’t released their data.”