As we navigate the complexities of the digital age, computational imaging has emerged as a pivotal discipline, integrating advanced algorithms and artificial intelligence to enhance image processing techniques. With the rapid proliferation of high-resolution imaging technologies, there is an increasing demand for efficient coding and representation techniques that can optimize storage, transmission, and analysis of visual data. This Collection aims to explore the intersection of computational imaging and AI, investigating how novel methodologies can transform the way we capture, process, and interpret images.

The significance of this research extends beyond technical advancements; it is crucial for various fields, including medical imaging, autonomous vehicles, and entertainment. Recent breakthroughs in machine learning and neural networks have enabled unprecedented levels of image enhancement and classification. Techniques such as deep learning-based compression and generative adversarial networks (GANs) have revolutionized how we approach image representation. As industries increasingly rely on image data, developing efficient coding techniques becomes essential for enhancing data throughput while minimizing resource consumption.

Looking ahead, the continued exploration of computational imaging holds immense potential for future innovations. We may witness the development of real-time image processing solutions that integrate seamlessly with augmented reality and virtual reality systems. Furthermore, as AI continues to evolve, we can expect smarter algorithms that not only compress and enhance images but also learn from user interactions, leading to more intuitive and personalized imaging experiences.

- Advanced compression techniques using deep learning

- Machine learning applications in medical imaging

- Efficient coding methods for high-resolution video

- Representation techniques in augmented and virtual reality

Researchers are invited to submit their work to this Collection, which aims to showcase innovative research that contributes to the advancements in computational imaging and efficient coding techniques.

The "Computational Imaging in the AI Era: Efficient Coding and Representation Techniques" Collection seeks contributions that explore the integration of artificial intelligence with computational imaging. This Collection welcomes research on advanced compression methods, machine learning applications in various imaging domains, and innovative representation techniques. By fostering collaboration and knowledge sharing, this Collection aims to advance the field of computational imaging and enhance our understanding of efficient coding in the digital age.

SDG 9: Industry, Innovation & Infrastructure

SDG 8: Decent Work and Economic Growth

Behavioral and social computing aims to understand human actions, interactions, and societal dynamics through computational methods. The advent of AI has ignited a transformative era for behavioral and social computing, creating unprecedented urgency to advance human-centric multimedia processing that harnesses synthetic intelligence for societal benefit. As multimedia data (video, audio, physiological signals, etc.) becomes the cornerstone of understanding human behavior, advanced processing techniques to capture nuanced aspects of human behavior in real-world settings are essential to decode complex social and individual patterns. This special issue focuses on the intersection of multimedia processing and behavioral/social computing, exploring novel methodologies, applications, and ethical considerations in leveraging multimodal data to model human behavior, emotions, social dynamics, and cultural trends. The special issue aims to pioneer next-generation multimedia processing methodologies for human-centric AI research that understand and respond to human behaviors and social dynamics, empowering researchers to tackle real-world challenges in human-AI interaction, healthcare, education, psychology, computational social sciences, business and marketing, and any other related areas.

The special issue seeks high-quality, original contributions from researchers, practitioners, and industry experts to present novel methodologies, real-world applications, comprehensive reviews, and experimental validations that leverage multimodal data to advance behavioral insights and social intelligence. Additionally, extended versions of selected high-quality papers from BESC2025, as well as notable conferences, such as MM, KDD, CIKM, AAAI, IJCAI, ICML, CVPR, ICCV, ECCV, ICIP, ICASSP, ICME, ICMR, ICLR, within the field will be invited to enrich the scope of this special issue.

The topics of interest for this special issue include, but are not limited to:

• Multimodal behavioral and social sensing and analytics

• Generative AI for synthetic behavioral and social data generation

• AI-driven large-scale social dynamics mining and discovery

• Human-AI collaboration for behavioral and social computing

• Human-centered affective computing

• Immersive multimedia solutions for training or therapy

• Conversational AI for social cue interpretation

• Ethical human-AI alignment in behavioral and social interventions

• Multimedia applications for healthcare, education, psychology, computational social sciences, business and marketing, etc.

All submitted papers will undergo rigorous peer-review by specialists in the field, with accepted papers set to be published as soon as they are ready.