http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
뇌파 기반의 감정 분류 및 맞춤 회화 생성에 관한 연구
이하윤(Hayun Lee),이태민(Taemin Lee) 한국디지털콘텐츠학회 2021 한국디지털콘텐츠학회논문지 Vol.22 No.10
Non-professionals who do not deal with much art content face various difficulties when producing content; Knowing exactly how you feel and how you express it in content. This paper proposes a system that solves the difficulties of non-experts in producing art content, especially painting. Customized painting is recommended based on the users emotion, and the input photo image is changed to the same style as the recommended painting. To this end, the most effective biometric information is used to predict emotions. It is recommended from a painting database that measures EEG to find their emotional state at a positive-negative ratio, and investigates the painting that best suits the found emotional state in advance. Deep learning-based style transfer techniques are applied to the recommended painting and input photo images. Finally, the recommended style of painting is copied to the input photo image as it is, so that a new painting image that reflects the users emotions can be created. This study has the advantage of being able to easily produce painting works without a painting assistance tool.
Vector engine 기반 NPU를 위한 grouped pattern-wise pruning
서수희(Soohee Seo),이하윤(Hayun Lee, Sangho Lee),이상호(Dongkun Shin),신동균(Dongkun S) 대한전자공학회 2023 대한전자공학회 학술대회 Vol.2023 No.6
Among the recent pruning techniques, it is conducted in various ways such as block-wise and channel-wise, and the accuracy is low because it is coarser than the relatively dense model. In addition, there is a disadvantage that hardware must be designed accordingly. To solve this problem, we propose an optimized pruning technique for VTA accelerator GEMM Core. Accuracy was 5.73% to 23.83% higher than block-wise pruned model. It is expected that there will be scalability that can be applied to various vector engine-base NPUs using this technique.