Presentation | 2014-01-24 Trend-Sensitive Hough Forests : Action Detection Method Using Voting Trends Kensho HARA, Takatsugu HIRAYAMA, Kenji MASE, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Action detection using the Hough voting approach can achieve robustness to occlusions because the approach casts votes for action classes and spatio-temporal action positions based on independent local features. However, each local feature has local information that tends to lead false votes. This paper uses a trend of the past voting to reduce false votes. We extend conventional Hough forests in order to use the trend. This extension can reduce false votes that conflict the trend. We called the proposed method Trend-Sensitive Hough Forests. In the experiment, we confirmed the decrease of false detection, so that the proposed method improved action detection accuracy compared with conventional Hough forests. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Action detection / Hough transform / Hough forests / Voting trends |
Paper # | PRMU2013-109,MVE2013-50 |
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) | Trend-Sensitive Hough Forests : Action Detection Method Using Voting Trends |
Sub Title (in English) | |
Keyword(1) | Action detection |
Keyword(2) | Hough transform |
Keyword(3) | Hough forests |
Keyword(4) | Voting trends |
1st Author's Name | Kensho HARA |
1st Author's Affiliation | Graduate School of Information Science, Nagoya University() |
2nd Author's Name | Takatsugu HIRAYAMA |
2nd Author's Affiliation | Graduate School of Information Science, Nagoya University |
3rd Author's Name | Kenji MASE |
3rd Author's Affiliation | Graduate School of Information Science, Nagoya University |
Date | 2014-01-24 |
Paper # | PRMU2013-109,MVE2013-50 |
Volume (vol) | vol.113 |
Number (no) | 403 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |