Search

Static machine learning methods in gesture recognition assume that training and test data come from the same underlying distribution. However, in real-world applications involving gesture recognition on wrist-worn devices, data distribution may change over time. We formulate this problem of adapting recognition models to new tasks, where new data patterns emerge, as open-world gesture recognition (OWGR). We propose leveraging continual learning to make machine learning models adaptive to new tasks without degrading performance on previously learned tasks. However, the exploration of parameters for questions around when and how to train and deploy recognition models requires time-consuming user studies and is sometimes impractical. To address this challenge, we propose a design engineering approach that enables offline analysis on a collected large-scale dataset with various parameters and compares different continual learning methods. Finally, design guidelines are provided to enhance the development of an open-world wrist-worn gesture recognition process.

Peiyang Ding,1,2 Teng Zhang,2,3 Yafei Li,1,2 Man Teng,2 Yaning Sun,2 Xiao Liu,2,4 Shujun Chai,2 Enmin Zhou,1 Qianyue Jin,2,5 Gaiping Zhang1,2,4,5 1College of Veterinary Medicine, Northwest A&F University, Yangling, 2Henan Provincial Key Laboratory of Animal Immunology, Henan Academy of Agricultural Sciences, 3College of Life Sciences, Henan Agricultural University, 4College of Animal Science and Veterinary Medicine, Henan Agricultural University, Zhengzhou, 5Jiangsu Co-innovation Center for Prevention and Control of Important Animal Infectious Diseases and Zoonoses, Yangzhou University, Yangzhou, People’s Republic of China Abstract: Recent advancements in biotechnology have enabled the rapid identification and subsequent expression of pathogenic microbial major antigens that induce protective immune responses. However, subunit vaccines have not been successfully commercialized mainly due to the lack of sufficient levels of neutralizing antibodies (NAs). High levels of NA rely on the efficient recognition and cross-linking of multiple neutralizing epitopes with B-cell receptors (BCRs). Nanoparticles are able to display coupled antigenic arrays at high density and provide multiple binding molecular scenarios with BCRs. The high-resolution antigenic structure makes it possible to accurately display stable neutralizing epitopes. Therefore, the development of a nanovaccine that orientationally displays neutralizing epitopes is a feasible strategy. To address this hypothesis, the capsid (Cap) protein of porcine circovirus type 2 as model antigen was conjugated to gold nanoparticles (AuNPs) through direct reaction of the mercapto group of the unique cysteines with AuNPs, rendering Cap-AuNPs to have neutralizing epitopes on outer surface and an immunodominant epitope buried within the inner surface. In vitro studies showed that AuNPs promoted the phagocytosis of Cap protein and NA levels were significantly improved, meanwhile antibody levels against the im

Peiyang Ding,1,2 Teng Zhang,2,3 Yafei Li,1,2 Man Teng,2 Yaning Sun,2 Xiao Liu,2,4 Shujun Chai,2 Enmin Zhou,1 Qianyue Jin,2,5 Gaiping Zhang1,2,4,5 1College of Veterinary Medicine, Northwest A&F University, Yangling, 2Henan Provincial Key Laboratory of Animal Immunology, Henan Academy of Agricultural Sciences, 3College of Life Sciences, Henan Agricultural University, 4College of Animal Science and Veterinary Medicine, Henan Agricultural University, Zhengzhou, 5Jiangsu Co-innovation Center for Prevention and Control of Important Animal Infectious Diseases and Zoonoses, Yangzhou University, Yangzhou, People’s Republic of China Abstract: Recent advancements in biotechnology have enabled the rapid identification and subsequent expression of pathogenic microbial major antigens that induce protective immune responses. However, subunit vaccines have not been successfully commercialized mainly due to the lack of sufficient levels of neutralizing antibodies (NAs). High levels of NA rely on the efficient recognition and cross-linking of multiple neutralizing epitopes with B-cell receptors (BCRs). Nanoparticles are able to display coupled antigenic arrays at high density and provide multiple binding molecular scenarios with BCRs. The high-resolution antigenic structure makes it possible to accurately display stable neutralizing epitopes. Therefore, the development of a nanovaccine that orientationally displays neutralizing epitopes is a feasible strategy. To address this hypothesis, the capsid (Cap) protein of porcine circovirus type 2 as model antigen was conjugated to gold nanoparticles (AuNPs) through direct reaction of the mercapto group of the unique cysteines with AuNPs, rendering Cap-AuNPs to have neutralizing epitopes on outer surface and an immunodominant epitope buried within the inner surface. In vitro studies showed that AuNPs promoted the phagocytosis of Cap protein and NA levels were significantly improved, meanwhile antibody levels against the im