Dr. Ruochen Li, BOHUA UHD Co., Ltd. , China.
Ruochen Li, PhD candidate at Macau University of Science and Technology, specializes in Artificial Intelligence with a focus on no-reference video quality assessment, cross-modal audio-visual retrieval, and image-based sound source localization. With expertise in cutting-edge AI technologies like PyTorch, TensorFlow, and MindSpore, Li has achieved groundbreaking research in video quality evaluation and audio-visual content correlation, earning recognition in top-tier journals. He has also received a prize in the National Artificial Intelligence Competition for his contributions to ultra-high-definition video processing. 
Publication Profile
Education and Experience
PhD in Artificial Intelligence (2021-2024), Macau University of Science and Technology.- Master’s in Control Engineering (2016-2019), Jiangsu University of Science and Technology.
- Supervisor: Associate Prof. Shuxia Ye.
- Bachelor’s in Control Engineering (2012-2016), Jiangsu University of Science and Technology.
Research Participant: National Ultra-High Definition Video Innovation Center.- Research Contributor: China Science and Technology Information Research Institute.
Suitability For The Award
Dr. Ruochen Li is an accomplished researcher specializing in artificial intelligence, video quality assessment, and audio-visual event retrieval. With a Ph.D. in Artificial Intelligence from Mauca University of Science and Technology and extensive expertise in PyTorch, TensorFlow, and MindSpore, Li has contributed significantly to advancing multimedia technologies. Their innovations include state-of-the-art datasets, algorithms like Reformer, and multimodal fusion techniques with applications in accessibility, entertainment, and surveillance. Recognized through high-impact publications and awards, including third prize in the National Artificial Intelligence Competition, Ruochen Li exemplifies excellence in research and innovation, making them a strong candidate for prestigious honors such as the Best Researcher Award.
Professional Development
Ruochen Li’s professional journey is defined by innovations in AI and deep learning. He developed the UHD-VQ5k dataset and proposed novel algorithms for ultra-high-definition video quality assessment, utilizing advanced models like Resformer. His work in audio-visual content analysis, featured in his doctoral dissertation, emphasizes the integration of audio-visual features using deep neural networks. As a key participant in national projects, he has contributed to cloud-based UHD video platforms and AI policy analysis. His collaborations and publications underscore his commitment to advancing AI research and applications. 
Research Focus
Ruochen Li’s research revolves around Artificial Intelligence applications in multimedia. His expertise spans no-reference video quality assessment, where he develops datasets and benchmarks for UHD video, to cross-modal audio-visual retrieval, enhancing machine understanding of multimodal content. His work also extends to image-based sound source localization, integrating audio-visual data for precise event detection. Through pioneering algorithms, Li bridges gaps between modalities, advancing the interplay of audio and video content in deep learning applications. His contributions drive progress in multimedia AI. 
Awards and Honors
Prize Winner: National Artificial Intelligence Competition.
CET-6 Certificate: Scored 490.- CET-4 Certificate: Scored 552.
Publication Top Notes
-
SgLFT: Semantic-guided Late Fusion Transformer for Video Corpus Moment Retrieval – Neurocomputing, 2024. 
-
Ultrahigh-definition Video Quality Assessment: A New Dataset and Benchmark – Neurocomputing, 2024,
-
TA2V: Text-Audio Guided Video Generation – IEEE Transactions on Multimedia, 2024, 
-
Cross-Modality Knowledge Calibration Network for Video Corpus Moment Retrieval – IEEE Transactions on Multimedia, 2024, 
-
Maximizing Mutual Information Inside Intra- and Inter-Modality for Audio-Visual Event Retrieval – International Journal of Multimedia Information Retrieval, 2023, 
