Gathering Effective Information for Real-Time Material Recognition
Material recognition is a fundamental problem in the field of computer vision. Material recognition is still challenging because of varying camera perspectives, light conditions, and illuminations. Feature learning or feature engineering helps build an important foundation for effective material rec...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9180292/ |
| _version_ | 1818446630194511872 |
|---|---|
| author | Hongbin Zhang Ziliang Jiang Qipeng Xiong Jinpeng Wu Tian Yuan Guangli Li Yiwang Huang Donghong Ji |
| author_facet | Hongbin Zhang Ziliang Jiang Qipeng Xiong Jinpeng Wu Tian Yuan Guangli Li Yiwang Huang Donghong Ji |
| author_sort | Hongbin Zhang |
| collection | DOAJ |
| description | Material recognition is a fundamental problem in the field of computer vision. Material recognition is still challenging because of varying camera perspectives, light conditions, and illuminations. Feature learning or feature engineering helps build an important foundation for effective material recognition. Most traditional and deep learning-based features usually point to the same or similar material semantics from diverse visual perspectives, indicating the implicit complementary information (or cross-modal semantics) among these heterogeneous features. However, only a few studies focus on mining the cross-modal semantics among heterogeneous image features, which can be used to boost the final recognition performance. To address this issue, we first improve the well-known multiset discriminant correlation analysis model to fully mine the cross-modal semantics among heterogeneous image features. Then, we propose a novel hierarchical multi-feature fusion (HMF<sup>2</sup>) model to gather effective information and create novel yet more effective and robust features. Finally, a general classifier is employed to train a new material recognition model. Experimental results demonstrate the simplicity, effectiveness, robustness, and efficiency of the HMF<sup>2</sup> model on two benchmark datasets. Furthermore, based on the HMF<sup>2</sup> model, we design an end-to-end online system for real-time material recognition. |
| first_indexed | 2024-12-14T19:50:46Z |
| format | Article |
| id | doaj.art-f0b0ae0ccd594afbb6cf132a65814d5d |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-14T19:50:46Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-f0b0ae0ccd594afbb6cf132a65814d5d2022-12-21T22:49:26ZengIEEEIEEE Access2169-35362020-01-01815951115952910.1109/ACCESS.2020.30203829180292Gathering Effective Information for Real-Time Material RecognitionHongbin Zhang0https://orcid.org/0000-0002-8375-0039Ziliang Jiang1Qipeng Xiong2https://orcid.org/0000-0002-0103-9084Jinpeng Wu3Tian Yuan4Guangli Li5Yiwang Huang6Donghong Ji7School of Software, East China Jiaotong University, Nanchang, ChinaSchool of Software, East China Jiaotong University, Nanchang, ChinaSchool of Software, East China Jiaotong University, Nanchang, ChinaSchool of Software, East China Jiaotong University, Nanchang, ChinaSchool of Information Engineering, East China Jiaotong University, Nanchang, ChinaSchool of Information Engineering, East China Jiaotong University, Nanchang, ChinaGuizhou Provincial Key Laboratory of Public Big Data, Guizhou University, Guiyang, ChinaSchool of Cyber Science and Engineering, Wuhan University, Wuhan, ChinaMaterial recognition is a fundamental problem in the field of computer vision. Material recognition is still challenging because of varying camera perspectives, light conditions, and illuminations. Feature learning or feature engineering helps build an important foundation for effective material recognition. Most traditional and deep learning-based features usually point to the same or similar material semantics from diverse visual perspectives, indicating the implicit complementary information (or cross-modal semantics) among these heterogeneous features. However, only a few studies focus on mining the cross-modal semantics among heterogeneous image features, which can be used to boost the final recognition performance. To address this issue, we first improve the well-known multiset discriminant correlation analysis model to fully mine the cross-modal semantics among heterogeneous image features. Then, we propose a novel hierarchical multi-feature fusion (HMF<sup>2</sup>) model to gather effective information and create novel yet more effective and robust features. Finally, a general classifier is employed to train a new material recognition model. Experimental results demonstrate the simplicity, effectiveness, robustness, and efficiency of the HMF<sup>2</sup> model on two benchmark datasets. Furthermore, based on the HMF<sup>2</sup> model, we design an end-to-end online system for real-time material recognition.https://ieeexplore.ieee.org/document/9180292/Material recognitionhierarchical multi-feature fusiondiscriminant correlation analysiscross-modal semantics |
| spellingShingle | Hongbin Zhang Ziliang Jiang Qipeng Xiong Jinpeng Wu Tian Yuan Guangli Li Yiwang Huang Donghong Ji Gathering Effective Information for Real-Time Material Recognition IEEE Access Material recognition hierarchical multi-feature fusion discriminant correlation analysis cross-modal semantics |
| title | Gathering Effective Information for Real-Time Material Recognition |
| title_full | Gathering Effective Information for Real-Time Material Recognition |
| title_fullStr | Gathering Effective Information for Real-Time Material Recognition |
| title_full_unstemmed | Gathering Effective Information for Real-Time Material Recognition |
| title_short | Gathering Effective Information for Real-Time Material Recognition |
| title_sort | gathering effective information for real time material recognition |
| topic | Material recognition hierarchical multi-feature fusion discriminant correlation analysis cross-modal semantics |
| url | https://ieeexplore.ieee.org/document/9180292/ |
| work_keys_str_mv | AT hongbinzhang gatheringeffectiveinformationforrealtimematerialrecognition AT ziliangjiang gatheringeffectiveinformationforrealtimematerialrecognition AT qipengxiong gatheringeffectiveinformationforrealtimematerialrecognition AT jinpengwu gatheringeffectiveinformationforrealtimematerialrecognition AT tianyuan gatheringeffectiveinformationforrealtimematerialrecognition AT guanglili gatheringeffectiveinformationforrealtimematerialrecognition AT yiwanghuang gatheringeffectiveinformationforrealtimematerialrecognition AT donghongji gatheringeffectiveinformationforrealtimematerialrecognition |