Continuous Human Action Recognition Using Depth-MHI-HOG and a Spotter Model
In this paper, we propose a new method for spotting and recognizing continuous human actions using a vision sensor. The method is comprised of depth-MHI-HOG (DMH), action modeling, action spotting, and recognition. First, to effectively separate the foreground from background, we propose a method ca...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2015-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/15/3/5197 |
| _version_ | 1811262987928862720 |
|---|---|
| author | Hyukmin Eum Changyong Yoon Heejin Lee Mignon Park |
| author_facet | Hyukmin Eum Changyong Yoon Heejin Lee Mignon Park |
| author_sort | Hyukmin Eum |
| collection | DOAJ |
| description | In this paper, we propose a new method for spotting and recognizing continuous human actions using a vision sensor. The method is comprised of depth-MHI-HOG (DMH), action modeling, action spotting, and recognition. First, to effectively separate the foreground from background, we propose a method called DMH. It includes a standard structure for segmenting images and extracting features by using depth information, MHI, and HOG. Second, action modeling is performed to model various actions using extracted features. The modeling of actions is performed by creating sequences of actions through k-means clustering; these sequences constitute HMM input. Third, a method of action spotting is proposed to filter meaningless actions from continuous actions and to identify precise start and end points of actions. By employing the spotter model, the proposed method improves action recognition performance. Finally, the proposed method recognizes actions based on start and end points. We evaluate recognition performance by employing the proposed method to obtain and compare probabilities by applying input sequences in action models and the spotter model. Through various experiments, we demonstrate that the proposed method is efficient for recognizing continuous human actions in real environments. |
| first_indexed | 2024-04-12T19:36:53Z |
| format | Article |
| id | doaj.art-ce4a22872a154b3bb31f125fb5a558dd |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-12T19:36:53Z |
| publishDate | 2015-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-ce4a22872a154b3bb31f125fb5a558dd2022-12-22T03:19:11ZengMDPI AGSensors1424-82202015-03-011535197522710.3390/s150305197s150305197Continuous Human Action Recognition Using Depth-MHI-HOG and a Spotter ModelHyukmin Eum0Changyong Yoon1Heejin Lee2Mignon Park3School of Electrical and Electronic Engineering, Yonsei University, 134 Shinchon-Dong, Seodaemun-Gu, Seoul 120-749, KoreaDepartment of Electrical Engineering, Suwon Science College, Hwaseong 445-742, KoreaDepartment of Electrical, Electronic and Control Engineering, Hankyong National University, Anseong 456-749, KoreaSchool of Electrical and Electronic Engineering, Yonsei University, 134 Shinchon-Dong, Seodaemun-Gu, Seoul 120-749, KoreaIn this paper, we propose a new method for spotting and recognizing continuous human actions using a vision sensor. The method is comprised of depth-MHI-HOG (DMH), action modeling, action spotting, and recognition. First, to effectively separate the foreground from background, we propose a method called DMH. It includes a standard structure for segmenting images and extracting features by using depth information, MHI, and HOG. Second, action modeling is performed to model various actions using extracted features. The modeling of actions is performed by creating sequences of actions through k-means clustering; these sequences constitute HMM input. Third, a method of action spotting is proposed to filter meaningless actions from continuous actions and to identify precise start and end points of actions. By employing the spotter model, the proposed method improves action recognition performance. Finally, the proposed method recognizes actions based on start and end points. We evaluate recognition performance by employing the proposed method to obtain and compare probabilities by applying input sequences in action models and the spotter model. Through various experiments, we demonstrate that the proposed method is efficient for recognizing continuous human actions in real environments.http://www.mdpi.com/1424-8220/15/3/5197continuous human action recognitiondepth-MHI-HOG (DMH)hidden Markov modelaction modelingaction spottingspotter model |
| spellingShingle | Hyukmin Eum Changyong Yoon Heejin Lee Mignon Park Continuous Human Action Recognition Using Depth-MHI-HOG and a Spotter Model Sensors continuous human action recognition depth-MHI-HOG (DMH) hidden Markov model action modeling action spotting spotter model |
| title | Continuous Human Action Recognition Using Depth-MHI-HOG and a Spotter Model |
| title_full | Continuous Human Action Recognition Using Depth-MHI-HOG and a Spotter Model |
| title_fullStr | Continuous Human Action Recognition Using Depth-MHI-HOG and a Spotter Model |
| title_full_unstemmed | Continuous Human Action Recognition Using Depth-MHI-HOG and a Spotter Model |
| title_short | Continuous Human Action Recognition Using Depth-MHI-HOG and a Spotter Model |
| title_sort | continuous human action recognition using depth mhi hog and a spotter model |
| topic | continuous human action recognition depth-MHI-HOG (DMH) hidden Markov model action modeling action spotting spotter model |
| url | http://www.mdpi.com/1424-8220/15/3/5197 |
| work_keys_str_mv | AT hyukmineum continuoushumanactionrecognitionusingdepthmhihogandaspottermodel AT changyongyoon continuoushumanactionrecognitionusingdepthmhihogandaspottermodel AT heejinlee continuoushumanactionrecognitionusingdepthmhihogandaspottermodel AT mignonpark continuoushumanactionrecognitionusingdepthmhihogandaspottermodel |