Video summarization has become one of the most effective solutions for quickly understanding a large amount of video data. Video properties such as importance, diversity, representativeness, and storyness have been wi...
详细信息
Context-aware systems(***)integrate cyber and physical space to provide adaptive functionalities in response to changes in *** context-aware systems is challenging due to the uncertain running ***,many input validatio...
详细信息
Context-aware systems(***)integrate cyber and physical space to provide adaptive functionalities in response to changes in *** context-aware systems is challenging due to the uncertain running ***,many input validation approaches have been proposed to protect context-aware systems from uncertainty and keep them executing ***,in contrast to context-aware systems'prevailing in physical environments,most of those academic solutions(83%)are purely evaluated in simulated *** this article,we study whether this evaluation setting could lead to biased *** build a testing platform,RM-Testing,based on DJI RoboMaster robot car,to conduct the physical-environment based *** select three up-to-date input validation approaches,and compare their performance in the simulated environment and in the physical *** experimental results show that all three approaches'performance in simulated environments(improving task success rate by 82%compared with the system without the support of input validation)does differ from their performance in a physical environment(improving the task success rate by 50%).We also recognize three factors(scenario setting,physical platform and environmental model)that affect the performance of input validation approaches,based on an execution model of the context-aware system.
In classification problems, the datasets often have many unrelated or redundant features. The unrelated or redundant features may deteriorate the performance of classifier. Feature selection (FS) is an effective appro...
详细信息
The person identification task aims to generate robust human representation embeddings. Traditional methods have achieved competitive results, but they face occlusion challenges, making it difficult to distinguish bet...
详细信息
The person identification task aims to generate robust human representation embeddings. Traditional methods have achieved competitive results, but they face occlusion challenges, making it difficult to distinguish between humans occluded by objects or other humans. This paper proposes Pose-to-Human (P2H), a pose guidance framework via Gram matrix and knowledge transfer mechanism to tackle occlusion problems in Person Re-identification to address this issue. Our framework comprises four modules: Human Encoder, Pose Encoder, Compact Modules (CPM), and the Gram Matrix Guiding (GMG). The Pose Encoder generates pose information used to guide the human embedding models, facilitating learning of other parts of the human anatomy, which enables the model to focus on the unoccluded parts of the body, thus mitigating the reliance on the data for part-to-part matching, which is the first limitation of previous works. Additionally, it can reduce the demand for general appearance information from the human encoder model. Our method achieves competitive results through extensive experiments on five datasets compared to state-of-the-art methods, making it a promising framework for addressing the Re-Identification task.
The proliferation of Unmanned Aerial Vehicles (UAVs) in spatio-temporal applications has given rise to the need for Drone-as-a-Service (DaaS) platforms that support diverse tasks, ranging from urban air mobility to pr...
详细信息
This article employs the Random Forest model to conduct predictive research on the food and media industries in the Chinese stock market. The data originates from the historical trading data of the Shanghai Stock Exch...
详细信息
Large language models (LLMs) have made significant progress in NLP. However, their ability to memorize, represent, and leverage commonsense knowledge has been a well-known pain point. In this paper, we specifically fo...
详细信息
LLMs (Large Language Models) usually interact with users in the form of dialogue and generate responses following their instructions, which naturally require dialogue comprehension abilities. However, dialogue compreh...
详细信息
Temporal knowledge graphs (TKGs), consisting of graph snapshots evolving over time, have attracted substantial research attention in various areas like recommender systems and relation networks. Although significant r...
详细信息
The multiplication of a sparse matrix with a dense vector is a vital operation in linear algebra, with applications in numerous contexts. After earlier research on FPGA acceleration of this operation had shown the pot...
详细信息
暂无评论