Interactive Known-Item Search Using Semantic Textual and Colour Modalities

Zhang, Zhenxing; Albatal, Rami; Gurrin, Cathal; Smeaton, Alan F.

doi:10.1007/978-3-319-14442-9_29

Cited by 4 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Out of the nine participating teams [3,5,11,22,27,35,37,39,51], six managed to score points during the competition. Further analysis of the logs showed that one of the three non-scoring teams managed to solve one of the tasks but submitted its data using a wrong format.…”

Section: Overview Of Resultsmentioning

confidence: 99%

Interactive video search tools: a detailed analysis of the video browser showdown 2015

Cobârzan

Schoeffmann

Bailer

et al. 2016

Multimed Tools Appl

View full text Add to dashboard Cite

Interactive video retrieval tools developed over the past few years are emerging as powerful alternatives to automatic retrieval approaches by giving the user more control as well as more responsibilities. Current research tries to identify the best combinations of image, audio and text features that combined with innovative UI design maximize the tools (2017) 76:5539-5571 performance. We present the last installment of the Video Browser Showdown 2015 which was held in conjunction with the International Conference on MultiMedia Modeling 2015 (MMM 2015) and has the stated aim of pushing for a better integration of the user into the search process. The setup of the competition including the used dataset and the presented tasks as well as the participating tools will be introduced . The performance of those tools will be thoroughly presented and analyzed. Interesting highlights will be marked and some predictions regarding the research focus within the field for the near future will be made.

show abstract

Section: Overview Of Resultsmentioning

confidence: 99%

Interactive video search tools: a detailed analysis of the video browser showdown 2015

Cobârzan

Schoeffmann

Bailer

et al. 2016

Multimed Tools Appl

View full text Add to dashboard Cite

show abstract

“…Such paradigm, nevertheless, is only valid for ad-hoc video search [2,31,38], with assumption of having abundant positive examples to be harvested for training. For knownitem search (KIS) [22,26,53] or video moment search [13,19,20], where typically only one search target per query, leveraging relevancy feedback to harvest positive examples for query refinement becomes inapplicable. Note that relevancy feedback can also be applied to glean examples visually similar to but not positive of a search target [5,16].…”

Section: Related Workmentioning

confidence: 99%

Interactive Video Corpus Moment Retrieval using Reinforcement Learning

Ngo

2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Known-item video search is effective with human-in-the-loop to interactively investigate the search result and refine the initial query. Nevertheless, when the first few pages of results are swamped with visually similar items, or the search target is hidden deep in the ranked list, finding the know-item target usually requires a long duration of browsing and result inspection. This paper tackles the problem by reinforcement learning, aiming to reach a search target within a few rounds of interaction by long-term learning from user feedbacks. Specifically, the system interactively plans for navigation path based on feedback and recommends a potential target that maximizes the long-term reward for user comment. We conduct experiments for the challenging task of video corpus moment retrieval (VCMR) to localize moments from a large video corpus. The experimental results on TVR and DiDeMo datasets verify that our proposed work is effective in retrieving the moments that are hidden deep inside the ranked lists of CONQUER and HERO, which are the state-of-the-art auto-search engines for VCMR. CCS CONCEPTS• Information systems → Video search; • Computing methodologies → Neural networks.

show abstract

“…Based on our experience of previous participation [2], we followed two main guidelines to design the interactive user interface for this system. Firstly, the interface should be simple and straightforward to allow users efficiently browse when dealing with a large amount of scenes.…”

Section: Browsing Interfacementioning

confidence: 99%

“…However, an overall analysis on these approaches reveals that most of them focusing on the idea of searching/filtering and browsing the results, instead of considering of using the high-level semantic concepts or low-level visual descriptors. Based on the wide range of approaches used by previous participants [2] in the competition, we made an observation that, the subjective and imprecise user expressions pose challenges for teams to provide appropriate tools to aid users to form suitable queries in order to make use of the advanced visual retrieval capabilities which have been embedded in these interactive systems. Only the SIRET team [3] gained success for visual retrieval tasks by making good use of position-color feature signatures and designed a special color picker interface to allow users to draw coloured regions when searching for the interested video clip.…”

Section: Introductionmentioning

confidence: 99%

“…Based on previous participants work [2], in this competition, we'll focus on visual content exploration and navigation technologies, and aimed to build an interactive user interface with the abilities of browsing large sets of video content quickly and effectively. To achieve this goal, we developed an interactive exploratory search system with faceted navigation [4] based on the outputs of visual content analysis, such as automatically extract semantic concepts, object labels, and object attributes from image/video content using various approaches.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Faceted Navigation for Browsing Large Video Collection

Zhang

Gurrin

et al. 2016

MultiMedia Modeling

Self Cite

View full text Add to dashboard Cite

Abstract. This paper presents a content-based interactive video browsing system to address the challenge in a live video search competition to find specific video clips from a large video collection under time constraints. Since the target of this evaluation forum is to evaluate and demonstrate the development of interactive video search tools, we do not need to consider if the most commonly used query-by-example or query-by-text approaches for large-scale image/video retrieval are appropriate in this scenario. In this paper, we describe an interactive video retrieval system which employs the concept filters and faceted navigation to aid users quickly and intuitively locate the interested content when browsing in a large video collections based on automatically extracted semantic concepts, object labels and attributes from video content.

show abstract

Interactive Known-Item Search Using Semantic Textual and Colour Modalities

Cited by 4 publications

References 9 publications

Interactive video search tools: a detailed analysis of the video browser showdown 2015

Interactive video search tools: a detailed analysis of the video browser showdown 2015

Interactive Video Corpus Moment Retrieval using Reinforcement Learning

Faceted Navigation for Browsing Large Video Collection

Contact Info

Product

Resources

About