Multi-Scale Proposal Regression Network for Temporal Action Proposal Generation
Temporal action detection, as a branch of video analysis, aims to locate the time points when the actions start and end, and classify the actions occurred in videos into correct categories. Generating high-quality proposals is a key step in temporal action detection task. In this paper, we introduce...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8788517/ |
| _version_ | 1818922078673305600 |
|---|---|
| author | Jingye Zheng Dihu Chen Haifeng Hu |
| author_facet | Jingye Zheng Dihu Chen Haifeng Hu |
| author_sort | Jingye Zheng |
| collection | DOAJ |
| description | Temporal action detection, as a branch of video analysis, aims to locate the time points when the actions start and end, and classify the actions occurred in videos into correct categories. Generating high-quality proposals is a key step in temporal action detection task. In this paper, we introduce a novel network, named multi-scale proposal regression network (MPRN), for temporal action proposal generation. First, we take encoding visual features as input and predict action scores for time points, in order to group them to generate rough proposals. Then, we regress the proposal's boundaries to obtain more precise proposals via our multi-scale proposal regression network. Compared with SSN and TURN, our multi-scale regression segments are characterized by flexible boundaries. Experiments show that 1) Our method is better than other proposal generation methods on THUMOS-14 dataset and ActivityNet-v1.3 dataset. 2) The effectiveness of our method is due to its own architecture, not the selection of visual feature encoders. 3) Our proposal generation method can generate temporal proposals for unseen action classes, which shows the good generalization ability of our proposal generation method. |
| first_indexed | 2024-12-20T01:47:49Z |
| format | Article |
| id | doaj.art-48701c9d945c4e979639a18b728774c5 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-20T01:47:49Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-48701c9d945c4e979639a18b728774c52022-12-21T19:57:42ZengIEEEIEEE Access2169-35362019-01-01718386018386810.1109/ACCESS.2019.29333608788517Multi-Scale Proposal Regression Network for Temporal Action Proposal GenerationJingye Zheng0Dihu Chen1https://orcid.org/0000-0001-5432-8149Haifeng Hu2https://orcid.org/0000-0002-4884-323XSchool of Electronic and Information Technology, Sun Yat-sen University, Guangzhou, ChinaSchool of Electronic and Information Technology, Sun Yat-sen University, Guangzhou, ChinaSchool of Electronic and Information Technology, Sun Yat-sen University, Guangzhou, ChinaTemporal action detection, as a branch of video analysis, aims to locate the time points when the actions start and end, and classify the actions occurred in videos into correct categories. Generating high-quality proposals is a key step in temporal action detection task. In this paper, we introduce a novel network, named multi-scale proposal regression network (MPRN), for temporal action proposal generation. First, we take encoding visual features as input and predict action scores for time points, in order to group them to generate rough proposals. Then, we regress the proposal's boundaries to obtain more precise proposals via our multi-scale proposal regression network. Compared with SSN and TURN, our multi-scale regression segments are characterized by flexible boundaries. Experiments show that 1) Our method is better than other proposal generation methods on THUMOS-14 dataset and ActivityNet-v1.3 dataset. 2) The effectiveness of our method is due to its own architecture, not the selection of visual feature encoders. 3) Our proposal generation method can generate temporal proposals for unseen action classes, which shows the good generalization ability of our proposal generation method.https://ieeexplore.ieee.org/document/8788517/Convolutional neural networktemporal action detectiontemporal action proposal generationvideo analysis |
| spellingShingle | Jingye Zheng Dihu Chen Haifeng Hu Multi-Scale Proposal Regression Network for Temporal Action Proposal Generation IEEE Access Convolutional neural network temporal action detection temporal action proposal generation video analysis |
| title | Multi-Scale Proposal Regression Network for Temporal Action Proposal Generation |
| title_full | Multi-Scale Proposal Regression Network for Temporal Action Proposal Generation |
| title_fullStr | Multi-Scale Proposal Regression Network for Temporal Action Proposal Generation |
| title_full_unstemmed | Multi-Scale Proposal Regression Network for Temporal Action Proposal Generation |
| title_short | Multi-Scale Proposal Regression Network for Temporal Action Proposal Generation |
| title_sort | multi scale proposal regression network for temporal action proposal generation |
| topic | Convolutional neural network temporal action detection temporal action proposal generation video analysis |
| url | https://ieeexplore.ieee.org/document/8788517/ |
| work_keys_str_mv | AT jingyezheng multiscaleproposalregressionnetworkfortemporalactionproposalgeneration AT dihuchen multiscaleproposalregressionnetworkfortemporalactionproposalgeneration AT haifenghu multiscaleproposalregressionnetworkfortemporalactionproposalgeneration |