首页|Research Conducted at Nanjing University of Posts and Telecommunications Has Pro vided New Information about Robotics (A Heterogeneous Attention Fusion Mechanism for the Cross-environment Scene Classification of the Home Service Robot)
Research Conducted at Nanjing University of Posts and Telecommunications Has Pro vided New Information about Robotics (A Heterogeneous Attention Fusion Mechanism for the Cross-environment Scene Classification of the Home Service Robot)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Robotics is the subjec t of a report. According to news originating from Jiangsu, People's Republic of China, by NewsRx correspondents, research stated, "There have been many methods to improve the capacity of scene classification of service robots. However, most of them are proposed from a technical standpoint but without reference to any c ognitive principle of the brain, and furthermore, from design to evaluation, the particularity of the robot task is still not fully considered, such as cross -e nvironment generalization, explicit semantic preservation and interpretation." Our news journalists obtained a quote from the research from the Nanjing Univers ity of Posts and Telecommunications, "Thus, the scene cognitive behavior of robo ts is far from humans, and their environmental adaptability is still poor. It is difficult to complete learning place concepts from discrete fragments and then continuously perceiving them with a limited view in unvisited spaces. Inspired b y the recent findings from neuroscience, an attention -based global and object a ttribute fusion mechanism (AGOFM for short) constructed by three parts is propos ed to overcome these deficiencies. In the global attribute part, a global featur e extractor and a sequence context extractor are used to generate the holistic f eature. The involved context integrates limited views to form an overall impress ion of a scene for guiding attention. In the object attribute part, a novel obje ct vector is proposed. It simultaneously involves the detected object quantity, category and confidence information, which are all related to the vector index a nd high-level semantics. In the attention generation part, two sorted top -X cha racteristics deriving from the above two parts are fed into a fully connected (F C) network with batch normalization to generate effective attention. The attenti on weights are then applied to the batch normalized global and object vectors re spectively, and subsequently, the two heterogeneous information are directly fus ed by another FC network to achieve scene classification. The policies for multi -learner fusion and frame rejection are also provided. Finally, a novel evaluat ion paradigm is proposed that the model is trained on a discrete prior dataset, and then the inference is tested on a traditional dataset and two robot view dat asets. This simulates the cross -environment situation."
JiangsuPeople's Republic of ChinaAsi aEmerging TechnologiesMachine LearningNano-robotRobotRoboticsNanjing University of Posts and Telecommunications
NETL
NSTL
