File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1145/2733373.2806374
- Scopus: eid_2-s2.0-84962786475
Supplementary
-
Citations:
- Scopus: 0
- Appears in Collections:
Conference Paper: Partially common-semantic pursuit for RGB-D object recognition
Title | Partially common-semantic pursuit for RGB-D object recognition |
---|---|
Authors | |
Keywords | Deep networks Feature learning RGB-D object recognition RICA |
Issue Date | 2015 |
Citation | MM 2015 - Proceedings of the 2015 ACM Multimedia Conference, 2015, p. 959-962 How to Cite? |
Abstract | For the RGB-D object recognition task, the robust and rich representations can boost the performance. Most works employ feature learning approaches to learn specific representation for the RGB and depth modalities independently, while some directly learn common property. Different from them, this paper proposes a novel supervised feature learning method for RGB-D object recognition, named Partially Common-Semantic Learning (PCSL), which jointly captures the complementary and consistency semantic information from RGB and depth modalities. The complementary information is revealed by the individual modality, while the consistency is exploited by both modalities simultaneously. In PCSL, Reconstruction Independent Component Analysis (RICA) is extended to integrate the supervised information and learn both of the complementary and partially shared common semantic information. The proposed approach is evaluated on two public RGB-D datasets and achieves better performance than several state-of-The-Art methods. |
Persistent Identifier | http://hdl.handle.net/10722/345212 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Jin, Lu | - |
dc.contributor.author | Li, Zechao | - |
dc.contributor.author | Shu, Xiangbo | - |
dc.contributor.author | Gao, Shenghua | - |
dc.contributor.author | Tang, Jinhui | - |
dc.date.accessioned | 2024-08-15T09:25:56Z | - |
dc.date.available | 2024-08-15T09:25:56Z | - |
dc.date.issued | 2015 | - |
dc.identifier.citation | MM 2015 - Proceedings of the 2015 ACM Multimedia Conference, 2015, p. 959-962 | - |
dc.identifier.uri | http://hdl.handle.net/10722/345212 | - |
dc.description.abstract | For the RGB-D object recognition task, the robust and rich representations can boost the performance. Most works employ feature learning approaches to learn specific representation for the RGB and depth modalities independently, while some directly learn common property. Different from them, this paper proposes a novel supervised feature learning method for RGB-D object recognition, named Partially Common-Semantic Learning (PCSL), which jointly captures the complementary and consistency semantic information from RGB and depth modalities. The complementary information is revealed by the individual modality, while the consistency is exploited by both modalities simultaneously. In PCSL, Reconstruction Independent Component Analysis (RICA) is extended to integrate the supervised information and learn both of the complementary and partially shared common semantic information. The proposed approach is evaluated on two public RGB-D datasets and achieves better performance than several state-of-The-Art methods. | - |
dc.language | eng | - |
dc.relation.ispartof | MM 2015 - Proceedings of the 2015 ACM Multimedia Conference | - |
dc.subject | Deep networks | - |
dc.subject | Feature learning | - |
dc.subject | RGB-D object recognition | - |
dc.subject | RICA | - |
dc.title | Partially common-semantic pursuit for RGB-D object recognition | - |
dc.type | Conference_Paper | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1145/2733373.2806374 | - |
dc.identifier.scopus | eid_2-s2.0-84962786475 | - |
dc.identifier.spage | 959 | - |
dc.identifier.epage | 962 | - |