Higher Education Category Submission

469 – 光語智囊

The project aims to enhance the liveliness of classroom teaching, improve students' acceptance of knowledge and interest in the classroom. We hope to make a generative AI-driven, voice-interactive holographic light field 3D teaching aid. Powered by generative AI, voice commands can be turned into scientifically accurate 3D models or animations in real time, pushing the boundaries of pre-made content. Voice interaction glasses-free 3D can more intuitively and vividly display geographical, chemical, and physical changes, such as topographic evolution and molecular chemical reactions. Instead of the traditional interpretation and teaching of knowledge based on pictures alone, this teaching aid allows teachers to manipulate complex knowledge visualisation through language, and students can deepen their understanding through spatial interaction. Although a large number of teams in society have tried to combine VR and AR with classroom teaching, it is undeniable that the long-term wearing comfort of head-mounted displays is still poor, and the visual dizziness is also strong. Therefore, we wanted to design a generative AI glasses-free 3D teaching aid that is comfortable to use and can stimulate students' initiative. This project belongs to the next generation of intelligent multimedia learning resources, and the design concept follows the "spatial proximity principle" in Meyer's multimedia learning resource design principles. Our technical principle is to use generative AI models in collaboration with professional engines to ensure as smooth and dynamic presentation as possible with low-latency square rendering. The application of generative AI holographic light field 3D teaching aids transforms abstract knowledge into operable three-dimensional objects, which greatly improves the efficiency of course comprehension. It will also open up a vertical market for light field technology in education and accelerate the implementation of generative AI in education applications.

Looking back at our primary, secondary, and sixth form learning experiences, classroom instruction is often confined to one-way explanations of established problem-solving methods by teachers. When it comes to spatial structure and abstract concepts, they only rely on two-dimensional images on presentation slides for presentation, which makes it difficult for many students to understand due to their lack of spatial imagination. This monolithic teaching method lacks interest and seriously affects students' motivation and effectiveness in learning. Based on this personal experience, we came up with the idea of developing smart teaching aids, hoping to change the limitations of traditional teaching through technological innovation. Surprisingly, when we searched the literature using the keywords "teaching efficiency" and "smart teaching aids", we found only a small number of relevant studies. This situation shows that research into educational tools based on generative AI is still in its infancy, and there is still a lot of room for exploration. To this end, we innovatively came up with the idea of a generative AI holographic light field 3D teaching aid. The generative AI holographic light field 3D teaching aids follow the heuristic principle in the teaching principles, and turn ideas into reality through speech via intelligent teaching aids, which fully improves the attractiveness of teaching content and stimulates students' curiosity and enthusiasm. This kind of intelligent teaching aid, which is based on the learner's learning motivation theory, is conducive to stimulating students' intrinsic motivation and driving learning with interest and curiosity. In addition, the product conforms to the ARCS model, a model that stimulates and sustains learners' motivation to learn, increases learning engagement through attention and relevance, and stimulates students' desire to explore through interactive 3D visualisations. So I think it has a lot of potential and significance in the era of AI education.

The technology and data resources used in the project include light field display technology, generative AI technology, voice interaction system, and educational content database. Hardware and supply chain resources include display device components, arithmetic support, as well as R&D funding and cooperation with educational institutions.
The market validation strategy has the following aspects.
1. Competitor analysis and differentiated positioning. Analyze existing products, clarify their advantages, and take different pricing to launch differentiated versions.
2. Market demand research. Through questionnaires and industry reports to assess the acceptability of the school and the budget that can be invested, combined with qualitative interviews, discussions with teachers and education managers to verify the ability of the product to solve practical problems.
3. Develop a minimum feasible product and collect experience data through education exhibitions or school pilots.
Core features of the solution include generative AI-driven dynamic content creation, immersive teaching and learning experience with light field display technology, voice interaction and classroom collaboration.
Core features of the solution include generative AI-driven dynamic content creation, immersive teaching and learning experience with light field display technology, voice interaction and classroom collaboration.
Key strategies to ensure a positive user experience include natural interaction design, content accuracy and real-time, privacy and ethical protection. Natural interaction design experience is ensured by using voice commands, on-screen visual cues, and touchable interactions; real-time and accurate content is achieved by low-latency rendering and utilizing an educational knowledge base to calibrate the knowledge graph; and priority control is set for the teacher to prevent technology from dominating the classroom, while anonymizing the data processing to protect privacy.
In order to ensure the effectiveness of the core functions, the project combines three functional dimensions to set five technical performance indicators: content generation latency, content accuracy, frame rate, visual coverage, and command recognition accuracy.

Null

The innovation and creativity of the project are as follows:
1. New definition of educational scenarios
The traditional teaching screen is a passive display carrier, while this project uses generative AI to transform abstract concepts (e.g. electromagnetic field, molecular bonding) into interactive spatial structures, and students actively construct cognition through multi-angle observation and voice questioning, so that the display device has a tutor-like ability to “understand the content of the teaching - respond to the user's needs - and dynamically optimize the presentation”.
2. Balance cost and usability
Light field display technology has long been limited by the high cost of film and television/medical fields, through AI-generated content to replace the pre-made CGI material, reduce the cost of the school to buy third-party 3D content library, and break the dilemma of “affordable equipment can not afford to use the content”.
3. Breakthrough in technology integration mode
Generative AI content generation (dynamics) + light field display (spatial sense) + voice interaction (natural interaction) are integrated into a single educational device, solving the problem of “functional fragmentation” of traditional education technology products: AI real-time generation of 3D content in line with the teaching schedule, combined with the light field display to provide explorability in the physical space, and voice interaction to realize the operation of zero cost of learning to form a closed loop of “content - presentation - interaction” closed loop.
There are two aspects about improving the effectiveness of responding to user challenges:
1. solving the core pain points of traditional education technology, including visualization of abstract concepts, maintenance of classroom attention, and democratization of teaching technology.
2. creating incremental value, from “replacing the blackboard” to “reinventing the classroom”, the light field screen makes “wrap-around teaching” possible.

To meet the growing demands of users, our strategy involves integrating generative artificial intelligence into our design. Generative AI features dynamic update mechanisms and adaptive architectural design, with its core strengths lying in continuous learning, iterative optimization, and resource scalability.
Generative AI employs data-driven continuous learning, regularly fine-tuning models with new data such as user feedback and updated knowledge bases through incremental training. Simultaneously, it performs online learning by absorbing real-time user interactions or demands to dynamically adjust models and deliver optimized responses.
Generative AI also supports multi-task learning to address diverse user needs. For example, in educational applications with light-field displays, the AI can simultaneously handle 3D model generation, voice interaction interpretation, and exercise grading. The system undergoes regular major version updates, incorporating new knowledge and functional enhancements based on user requirements.
To ensure environmental sustainability, light-field displays may eventually replace traditional projectors and screens. Given their longer update cycles, light-field technology reduces long-term device replacement costs. To address escalating user demands, our primary solution is update-driven optimization, with future AI systems potentially evolving into self-improving architectures.
This design not only lowers costs but also embraces eco-friendly principles.

The proposed solution addresses key challenges such as uneven distribution of educational resources, difficulties in special education, and monotonous teaching methods.
In remote areas, the lack of high-quality teachers and laboratory equipment is mitigated by light-field 3D courseware and AI instructors, which break geographical barriers and democratize access to high-risk, high-cost content like anatomy and chemistry experiments. For visually or hearing-impaired students, traditional teaching methods often fall short. However, voice interaction + tactile feedback light-field models (e.g., touchable geometric shapes) enable multimodal, barrier-free learning. Meanwhile, the low engagement caused by rote learning is countered by AI-generated personalized interactive scenarios (e.g., role-playing historical events), significantly enhancing immersion. These use cases collectively advance educational equity.
To respond to community needs, we implement the following mechanisms:
- Technically: Monthly analysis of user voice logs to cluster high-frequency requests, prioritizing updates to AI modules.
- Organizationally: A community co-governance committee—comprising teachers, NGO representatives, and individuals with disabilities—votes quarterly on development priorities.

Personal Information Collection Statement (PICS):
1. The personal data collected in this form will be used for activity-organizing, record keeping and reporting only. The collected personal data will be purged within 6 years after the event.
2. Please note that it is obligatory to provide the personal data required.
3. Your personal data collected will be kept by the LTTC and will not be transferred to outside parties.
4. You have the right to request access to and correction of information held by us about you. If you wish to access or correct your personal data, please contact our staff at lttc@eduhk.hk.
5. The University’s Privacy Policy Statement can be access at https://www.eduhk.hk/en/privacy-policy.

• I have read and agree to the competition rules and privacy policy.