Chenyuan Qu

• 28/02/2024: 🎉 Our paper "360 + x: A Panoptic Multi-modal Scene Understanding Dataset" has been accepted by CVPR2024!
• 09/02/2024: 📝 Grad to accept the invitation to serve as ACM MM2024 reviewer!
• 14/07/2023: 🎉 Our paper "Multi-view Self-supervised Disentanglement for General Image Denoising" has been accepted by ICCV2023!

Full-Stack Engineer (Machine Learning)

• Designed and developed a recommendation system and search engine for Vieunite Application that leveraged biophilic information in addition to regular metrics. The system achieved a 300% increase in click-through and purchase rates, resulting in a significant boost in revenue for the company.
• Solely completed the Vieutopia back-end, an AI-generated artwork platform that utilizes text-to-image and image-to-image models. Optimized and trained the models to produce high-quality, aesthetically pleasing artwork, and implemented Knowledge Distillation to make inference faster and more efficient.
• Designed the security architecture for Vieutopia to ensure the protection of user data and prevent unauthorized access to the platform, which has now garnered over 200,000 users as of March 2023.

Research Assistant

• Conducted research on the interpretable modelling of hydrological models using machine learning techniques to improve accuracy while clearly describing their physical processes.
• Utilized neural networks to search for more appropriate model connections to find unknown underlying patterns.
• Collaborated with a team of researchers to analyze and interpret data and to develop innovative solutions for complex problems in the field of hydrology.

Algorithm Engineer - Intern

• Developed a visual customer service anchor using machine learning to synchronize mouth movements with emotions and body language, particularly in the Chinese language context, and aligned with Asian pronunciation habits.
• Optimized the customer service anchor model for mobile applications and collaborated with cross-functional teams to deploy the project in the China Mobile mobile app.
• Collaborated with a team of researchers to analyze and interpret data and to develop innovative solutions for complex problems in the field of hydrology.

University of Birmingham

University of Southampton

Loughborough Grammar School