Convolutional Hierarchical Attention Network for Query-Focused Video Summarization
نویسندگان
چکیده
منابع مشابه
Query-focused Video Summarization
We introduce a new approach to video summarization based on a user query. We introduce a novel dataset and explain a baseline algorithm that was used to test our set. Our set has been gathered for 4 different videos with summaries for 60 different query pairs per video. We show that our dataset can be used to train a variety of supervised algorithms for the task of video-
متن کاملQuery-Focused Extractive Video Summarization
Section A presents the list of concepts used in our experiments for UTE and TV episodes datasets. Section B studies the effect of ground set size on the performance of our proposed method, SH-DPP. Section C presents the training and optimization details of SH-DPP. Section D studies the lower dimensions of the model parameters W and V. Section E shows a few system generated video summaries in fu...
متن کاملQuery-Chain Focused Summarization
Update summarization is a form of multidocument summarization where a document set must be summarized in the context of other documents assumed to be known. Efficient update summarization must focus on identifying new information and avoiding repetition of known information. In Query-focused summarization, the task is to produce a summary as an answer to a given query. We introduce a new task, ...
متن کاملBayesian Query-Focused Summarization
We present BAYESUM (for “Bayesian summarization”), a model for sentence extraction in query-focused summarization. BAYESUM leverages the common case in which multiple documents are relevant to a single query. Using these documents as reinforcement for query terms, BAYESUM is not afflicted by the paucity of information in short queries. We show that approximate inference in BAYESUM is possible o...
متن کاملSupplementary Material for Query-Focused Video Summarization: Dataset, Evaluation, and A Memory Network Based Approach
In this section, we thoroughly describe the process of generating the queries from dense concept annotations. While users often input free text to query videos through search engines, we simulate the real scenarios and construct the queries using the dense concept annotations we have collected (cf. Section 3.2 in the main text) to ease benchmarking different algorithms. By processing the dense ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Proceedings of the AAAI Conference on Artificial Intelligence
سال: 2020
ISSN: 2374-3468,2159-5399
DOI: 10.1609/aaai.v34i07.6929