| Presentation | 2014-01-24 Instance Search with Exponential BM25 Masaya MURATA, Hidehisa NAGANO, Kunio KASHINO, Shin'ichi SATOH, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We propose a basic video retrieval and ranking method for instance search task. The task requires a system to search videos in which an instance (a specific person, object, or place) really appears. The instance search is one of the key tasks proposed by the international community on video retrieval research called TRECVID and it has been intensively discussed by the participating community people since 2010. Ameliorating the BM25 (Best Match 25), a state-of-the-art probabilistic retrieval method in the text retrieval field, for image/video retrieval tasks, we have created a new method (named the exponential BM25) and applied it to the instance search tests to find that the performance has recorded the high search accuracy. In this paper, we describe our methodology and show the experimental evaluation results using a series of the TRECVID dataset. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Video retrieval / Instance search / probabilistic retrieval method / TRECVID / exponential BM25 |
| Paper # | PRMU2013-111,MVE2013-52 |
| Date of Issue |
| Conference Information | |
| Committee | MVE |
|---|---|
| Conference Date | 2014/1/16(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Media Experience and Virtual Environment (MVE) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Instance Search with Exponential BM25 |
| Sub Title (in English) | |
| Keyword(1) | Video retrieval |
| Keyword(2) | Instance search |
| Keyword(3) | probabilistic retrieval method |
| Keyword(4) | TRECVID |
| Keyword(5) | exponential BM25 |
| 1st Author's Name | Masaya MURATA |
| 1st Author's Affiliation | NTT Communication Science Laboratories() |
| 2nd Author's Name | Hidehisa NAGANO |
| 2nd Author's Affiliation | NTT Communication Science Laboratories |
| 3rd Author's Name | Kunio KASHINO |
| 3rd Author's Affiliation | NTT Communication Science Laboratories:National Institute of Informatics |
| 4th Author's Name | Shin'ichi SATOH |
| 4th Author's Affiliation | National Institute of Informatics |
| Date | 2014-01-24 |
| Paper # | PRMU2013-111,MVE2013-52 |
| Volume (vol) | vol.113 |
| Number (no) | 403 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |