Programs

Chinese Language and Literature

The Department of Chinese Language and Literature at Shenzhen University, established in 1984 with support from Peking University, offers a comprehensive program focused on the study of Chinese language and literature. It emphasizes integrating general education, specialized knowledge, and practical application to cultivate students' learning, innovation, and practical abilities. The department is committed to serving societal development and has consistently improved its teaching quality and research output. It offers undergraduate programs and postgraduate degrees at Master's and Doctoral levels in various specialized fields within Chinese language and literature.

Clinical Research Program in Shenzhen

This program focuses on advancing clinical research capabilities in Shenzhen, China. It aims to identify and address the strengths and weaknesses in conducting high-quality clinical research within the region. By understanding the local landscape, the program seeks to propose policy recommendations and foster innovation in medical research. The study involved participants from hospitals, pharmaceutical/medical device companies, and government agencies. It utilized qualitative methods, including focus groups and individual interviews, to gather comprehensive insights. The analysis focused on identifying key factors that drive and challenge clinical research efforts in Shenzhen.

Computer Front-End Technology

Computer Science and Software Engineering

The College of Computer Science and Software Engineering at Shenzhen University was established in 2008, building upon earlier computer science education initiatives dating back to 1983. It has rapidly gained recognition, with its Computer Science and Engineering discipline ranked among the top globally and within China. The college is dedicated to advancing the fields of computer science and software engineering, fostering a strong research environment, and producing highly skilled graduates. The college is equipped with state-of-the-art facilities, including national and provincial-level laboratories focusing on areas like Big Data System Computing Technology and Artificial Intelligence. With a significant research output and funding, the faculty is actively engaged in cutting-edge projects. The college has a distinguished faculty, including members of prestigious academies, and has produced notable alumni, such as Ma Huateng, the founder of Tencent.

Controllable Video Generation: A Survey

This survey provides a comprehensive overview of controllable video generation, a rapidly developing subfield of AI-generated content (AIGC). As AI models become more sophisticated, there's a growing need for methods that allow users to precisely control video output to match their intent. Current text-to-video models often fall short because text prompts alone are insufficient for complex or fine-grained requests. To address this, researchers are integrating non-textual conditions, such as camera motion or human pose, into existing video generation models to enable more accurate and flexible video synthesis. The survey systematically reviews the theoretical foundations and recent advancements in controllable video generation. It covers key concepts, common open-source video generation models, and delves into control mechanisms within diffusion models. Methods are categorized based on the types of control signals used, including single-condition, multi-condition, and universal controllable generation. The goal is to offer a clear understanding of the field's progress and future directions.

Fine-grained Textual Inversion Network for Zero-Shot Composed Image Retrieval (FTI4CIR)

This research introduces a novel method called the Fine-grained Textual Inversion Network (FTI4CIR) designed for Zero-Shot Composed Image Retrieval (ZS-CIR). Composed Image Retrieval allows users to search for images using a combination of a reference image and a text description that specifies desired modifications. ZS-CIR aims to achieve this without requiring pre-annotated image-text triplets, addressing the challenge of expensive data annotation. The FTI4CIR network addresses limitations in existing ZS-CIR approaches that use a single pseudo-word token to represent an image. This can be too coarse to capture an image's full content. FTI4CIR proposes a more comprehensive approach by mapping an image into multiple textual representations: a subject-oriented pseudo-word token and several attribute-oriented pseudo-word tokens. This fine-grained mapping aims to more accurately represent the image in a textual format. Furthermore, FTI4CIR incorporates a tri-wise caption-based semantic regularization. This component works to align these fine-grained pseudo-word tokens with real-word token embeddings, guided by image captions generated using the BLIP model. The effectiveness of this method has been demonstrated through extensive experiments on three benchmark datasets, showing superior performance.

Language Program

Shenzhen University offers a comprehensive Language Program for international students interested in learning Chinese language and culture. This program is ideal for those seeking to improve their Mandarin skills for academic or personal enrichment. The College of International Exchange facilitates these language programs, providing a pathway for students to engage with the Chinese language in an immersive environment. Students can choose from various course durations and focuses, including long-term programs and specialized short-term options.

Metaverse for Healthcare: Technologies, Challenges, and Vision

This paper explores the transformative potential of the metaverse in revolutionizing healthcare services. As living conditions improve, so do the demands on medical and healthcare systems. Current challenges, including shortages of medical resources, inefficient treatment processes, and limitations in medical technology, can be addressed through metaverse integration. The metaverse offers a dynamic system that bridges virtual and physical realities, fostering collaboration among intelligent agents in the digital realm. This allows for the abstraction and modeling of real-world healthcare problems, enabling computational experiments and the development of strategies that can dynamically guide actions in the physical world. The outcomes and data from real-world applications continuously refine the virtual models, creating a feedback loop for ongoing improvement. The research presented summarizes the current state of metaverse applications in various healthcare scenarios. It highlights key technologies, identifies significant challenges that need to be overcome, and outlines a future vision for this rapidly evolving field. The aim is to stimulate further research and innovation in leveraging the metaverse to enhance the quality and accessibility of medical services.

Offline Long-term Chinese Program

PhD Computer Science

TinyML: Enabling of Inference Deep Learning Models on Ultra-Low-Power IoT Edge Devices for AI Applications

The Internet of Things (IoT) is rapidly expanding, with billions of devices collecting data about their environments. While these devices are becoming more intelligent through Machine Learning (ML), their limited resources pose a challenge for complex Deep Learning (DL) models. Sending all data to the cloud for processing can lead to delays, privacy concerns, and increased communication costs. Tiny Machine Learning (TinyML) offers a solution by enabling the local processing of data and the inference of ML models directly on resource-constrained devices like microcontrollers. This paper reviews the current state of TinyML, analyzing the types of ML models used, datasets, and device characteristics to identify development needs and future directions.