This project is set against the backdrop of "Chinese bed-curtain culture": research data shows that in certain regions, more than 80% of Chinese undergraduate dormitory beds use bed curtains. However, I have found that the enclosed micro-environment formed after closing the bed curtain at night has a clear impact on CO2 concentration, temperature and humidity, and that students show noticeably negative health effects after getting up the next morning...

• 1. A CO2 concentration, temperature and humidity monitor that keeps historical records and guarantees accuracy.
• 2. Ideally, at least 10 participants are needed, meeting the following requirements: Chinese university students living on campus, falling asleep between 23:00 and 1:00, getting up between 6:30 and 9:30, for at least 4 weeks; from Monday to Friday each week, wear a sleep-quality tracking wristband while sleeping, place an air-quality monitoring device inside the bed curtain, and complete a short questionnaire after getting up the next morning.

People can rank all kinds of things — open-source UI component websites, the strength of game characters, the popularity of anime characters, and so on. The ultimate goal is to achieve an effect similar to the Guinness World Records: just as people think of Guinness World Records when they want to find "the most XX", they will think of this site when they want to find anything ranking-related. It could be an interesting website, but technology alone cannot achieve this effect — promotion, operations and other factors are all essential to reaching the goal.

• 1. Collaborators for programming.
• 2. Collaborators for business.

This project aims to solve the dilemma of not knowing how to take a photo. By building a dataset of photographs with annotations (including photographic elements such as subject, composition and camera parameters) and training a model on this dataset, the goal is to generate annotations from an input scene image. Similar images are then retrieved from the database based on the annotations to provide photographic ideas.

• 1. Collaborators for data collection, data annotation and dataset construction.
• 2. Collaborators for model training.
• 3. Collaborators for mobile app development and App Store launch.