Instance-Level Future Motion Estimation in a Single Image Based on Ordinal Regression and Semi-Supervised Domain Adaptation
A novel algorithm to estimate instance-level future motion (FM) in a single image is proposed in this paper. First, the FM of an instance is defined with its direction, speed, and action classes. Then, a deep neural network, called FM-Net, is developed to determine the FM of the instance. More speci...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9121271/ |
| _version_ | 1819175864012636160 |
|---|---|
| author | Kyung-Rae Kim Yeong Jun Koh Chang-Su Kim |
| author_facet | Kyung-Rae Kim Yeong Jun Koh Chang-Su Kim |
| author_sort | Kyung-Rae Kim |
| collection | DOAJ |
| description | A novel algorithm to estimate instance-level future motion (FM) in a single image is proposed in this paper. First, the FM of an instance is defined with its direction, speed, and action classes. Then, a deep neural network, called FM-Net, is developed to determine the FM of the instance. More specifically, the multi-context pooling layer is proposed to exploit both object and global context features, and the cyclic ordinal regression scheme is developed using binary classifiers for effective FM classification. Also, the proposed FM-Net is trained in a semi-supervised domain adaptation setting to obtain reliable FM estimation results, even when a source domain in the training process and a target domain in the inference process are different. Extensive experimental results demonstrate that the proposed algorithm provides remarkable performance and thus can be used effectively for computer vision applications, including single object tracking, multiple object tracking, and crowd analysis. Furthermore, the FM dataset, collected from diverse sources and annotated manually, is released as a benchmark for single-image FM estimation. |
| first_indexed | 2024-12-22T21:01:38Z |
| format | Article |
| id | doaj.art-31e61ebb00244173a2103bd67599a746 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T21:01:38Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-31e61ebb00244173a2103bd67599a7462022-12-21T18:12:49ZengIEEEIEEE Access2169-35362020-01-01811508911510810.1109/ACCESS.2020.30037519121271Instance-Level Future Motion Estimation in a Single Image Based on Ordinal Regression and Semi-Supervised Domain AdaptationKyung-Rae Kim0Yeong Jun Koh1Chang-Su Kim2https://orcid.org/0000-0002-4276-1831School of Electrical Engineering, Korea University, Seoul, South KoreaDepartment of Computer Science and Engineering, Chungnam National University, Daejeon, South KoreaDepartment of Computer Science and Engineering, Chungnam National University, Daejeon, South KoreaA novel algorithm to estimate instance-level future motion (FM) in a single image is proposed in this paper. First, the FM of an instance is defined with its direction, speed, and action classes. Then, a deep neural network, called FM-Net, is developed to determine the FM of the instance. More specifically, the multi-context pooling layer is proposed to exploit both object and global context features, and the cyclic ordinal regression scheme is developed using binary classifiers for effective FM classification. Also, the proposed FM-Net is trained in a semi-supervised domain adaptation setting to obtain reliable FM estimation results, even when a source domain in the training process and a target domain in the inference process are different. Extensive experimental results demonstrate that the proposed algorithm provides remarkable performance and thus can be used effectively for computer vision applications, including single object tracking, multiple object tracking, and crowd analysis. Furthermore, the FM dataset, collected from diverse sources and annotated manually, is released as a benchmark for single-image FM estimation.https://ieeexplore.ieee.org/document/9121271/Future motion estimationcyclic ordinal regressionsemi-supervised domain adaptation |
| spellingShingle | Kyung-Rae Kim Yeong Jun Koh Chang-Su Kim Instance-Level Future Motion Estimation in a Single Image Based on Ordinal Regression and Semi-Supervised Domain Adaptation IEEE Access Future motion estimation cyclic ordinal regression semi-supervised domain adaptation |
| title | Instance-Level Future Motion Estimation in a Single Image Based on Ordinal Regression and Semi-Supervised Domain Adaptation |
| title_full | Instance-Level Future Motion Estimation in a Single Image Based on Ordinal Regression and Semi-Supervised Domain Adaptation |
| title_fullStr | Instance-Level Future Motion Estimation in a Single Image Based on Ordinal Regression and Semi-Supervised Domain Adaptation |
| title_full_unstemmed | Instance-Level Future Motion Estimation in a Single Image Based on Ordinal Regression and Semi-Supervised Domain Adaptation |
| title_short | Instance-Level Future Motion Estimation in a Single Image Based on Ordinal Regression and Semi-Supervised Domain Adaptation |
| title_sort | instance level future motion estimation in a single image based on ordinal regression and semi supervised domain adaptation |
| topic | Future motion estimation cyclic ordinal regression semi-supervised domain adaptation |
| url | https://ieeexplore.ieee.org/document/9121271/ |
| work_keys_str_mv | AT kyungraekim instancelevelfuturemotionestimationinasingleimagebasedonordinalregressionandsemisuperviseddomainadaptation AT yeongjunkoh instancelevelfuturemotionestimationinasingleimagebasedonordinalregressionandsemisuperviseddomainadaptation AT changsukim instancelevelfuturemotionestimationinasingleimagebasedonordinalregressionandsemisuperviseddomainadaptation |