Generative adversarial networks (GANs) have demonstrated great success in generating various visual content. However, images generated by existing GANs are often of attributes (e.g., smiling expression) learned from one image domain. As a result, generating images of multiple attributes requires many real samples possessing multiple attributes which are very resource expensive to be collected. In this paper, we propose a novel GAN, namely Intersect-GAN, to learn multiple attributes from different image domains through an intersecting architecture. For example, given two image domains X 1 and X 2 with certain attributes, the intersection X 1 ∩ X 2 denotes a new domain where images possess the attributes from both X 1 and X 2 domains. The proposed IntersectGAN consists of two discriminators D 1 and D 2 to distinguish between generated and real samples of different domains, and three generators where the intersection generator is trained against both discriminators. And an overall adversarial loss function is defined over three generators. As a result, our proposed IntersectGAN can be trained on multiple domains of which each presents one specific attribute, and eventually eliminates the need of real sample images simultaneously possessing multiple attributes. By using the CelebFaces Attributes dataset, our proposed IntersectGAN is able to produce high quality face images possessing multiple attributes (e.g., a face with black hair and a smiling expression). Both qualitative and quantitative evaluations are conducted to compare our proposed IntersectGAN with other baseline methods. Besides, several different applications of IntersectGAN have been explored with promising results. CCS CONCEPTS• Computing methodologies → Computer vision tasks; Learning paradigms.

To avoid the spread of COVID-19, China has implemented strict lockdown policies and control measures, resulting in a dramatic decrease in air pollution and improved air quality. In this study, the air quality model WRF-Chem and the latest MEIC2019 and MEIC2020 anthropogenic emission inventories were used to simulate the air quality during the COVID-19 lockdown in 2020 and the same period in 2019. By designing different emission scenarios, this study explored the impact of the COVID-19 lockdown on the concentration of air pollutants emitted by different sectors (industrial sector and transportation sector) in Nanjing for the first time. The results indicate that influenced by the COVID-19 lockdown policies, compared with the same period in 2019, the concentrations of PM2.5, PM10, and NO2 in Nanjing decreased by 15%, 17.1%, and 20.3%, respectively, while the concentration of O3 increased by 45.1% in comparison; the concentrations of PM2.5, PM10 and NO2 emitted by industrial sector decreased by 30.7%, 30.8% and 14.0% respectively; the concentrations of PM2.5, PM10 and NO2 emitted by transportation sector decreased by 15.6%, 15.7% and 26.2% respectively. The COVID-19 lockdown has a greater impact on the concentrations of PM2.5 and PM10 emitted by the industrial sector, while the impact on air pollutants emitted by the transportation sector is more reflected in the concentration of NO2. This study provides some theoretical basis for the treatment of air pollutants in different departments in Nanjing.