Face Attributes Classification (FAC) is an important task in computer vision, aiming to predict the facial attributes of a given image. However, the value of mid-level feature information and the correlation between face attributes are always ignored by deep learning-based FAC methods. In order to solve these problems, we propose a novel and effective Multitask CNN architecture. Instead of predicting all 40 attributes together, an attribute grouping strategy is proposed to divide the 40 attributes into 8 task groups correlatively. Meanwhile, through the Fusion Layer, mid-level deep representations are fused into the original feature representations to jointly predict the face attributes. Furthermore, the Taskunique Attention Modules can help learn more task-specific feature representations, obtaining higher FAC accuracy. Extensive experiments on the CelebA dataset demonstrate that our method outperforms state-of-the-art FAC methods.
In this Letter, the duty cycle generation scheme is presented and implemented, which can improve the input performances of the three‐phase single‐stage full‐bridge power factor correction (PFC) converter. The analysis and discussion is aiming at the operation of the PFC converter in balanced grid condition. On the basis of the instantaneous expression of the three‐phase input current, the ideal duty cycle of each phase is obtained. To achieve a good PF for each phase, a simplified duty cycle generation scheme is presented and implemented in the three‐phase PFC converter. The simulating and experimental results based on a 500 W prototype show that after adoption of the presented duty cycle generation scheme, a higher PF has been achieved for each phase of the three‐phase PFC converter.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.