Image and video processing technologies have significant importance in the development of modern society. Some important applications include security surveillance, medical diagnosis, autonomous driving, and smart homes. Traditionally, computers have had difficulty processing non-structured signals such as images and videos. In recent years, with the support of theories in neuroscience, a new generation of intelligent computation technologies based on neurons and neural networks has been developed. Thanks to their bio-inspired structure, they exhibit remarkably high performance in processing non-structured signals. For example, the introduction of the concept of receptive fields in convolutional neural networks has significantly improved the performance of traditional neural networks in processing non-structured information. However, although neuroscience theory has provided bio-inspired performance advantages for artificial intelligence, existing neuron models and neural network structures still fall far short of those in the brains of animals and humans. From this perspective, image and video intelligent processing technologies still require continued advancement in neuroscience theory.
This Research Topic encourages further development of intelligent methods and techniques for image and video processing based on neuroscience theory. This can be achieved by optimizing classical neural networks or building new neural networks by considering some known knowledge, such as designing more realistic neuron models and network hierarchy structures. There are three possible forms of such development: 1) using existing neuroscience theory to further advance methods for processing non-structured signals; 2) gaining inspiration from artificial intelligence systems to guide and deepen research in visual neuroscience; and 3) independently advancing theoretical cognition from both the fields of neuroscience and artificial intelligence.
This Research Topic aims to enhance researchers' understanding of existing artificial intelligence technologies and promote the deep integration and exploration of the development of neuroscience and intelligent visual technology. We hope that neuroscience can guide the design of more powerful biomimetic systems to better serve modern society and humanity. And we also hope that related research brings up promise for advancing our understanding of the brain and developing innovative intelligent technologies. Therefore, we are seeking original research papers on topics including, but not limited to:
Functional artificial neuron model
Brain-like system
Computer-aided diagnosis
Small-data driven learning
Image classification
Image object recognition
Image generation
Video classification
Motion detection
Video object tracking
Video action recognition
Image and video processing technologies have significant importance in the development of modern society. Some important applications include security surveillance, medical diagnosis, autonomous driving, and smart homes. Traditionally, computers have had difficulty processing non-structured signals such as images and videos. In recent years, with the support of theories in neuroscience, a new generation of intelligent computation technologies based on neurons and neural networks has been developed. Thanks to their bio-inspired structure, they exhibit remarkably high performance in processing non-structured signals. For example, the introduction of the concept of receptive fields in convolutional neural networks has significantly improved the performance of traditional neural networks in processing non-structured information. However, although neuroscience theory has provided bio-inspired performance advantages for artificial intelligence, existing neuron models and neural network structures still fall far short of those in the brains of animals and humans. From this perspective, image and video intelligent processing technologies still require continued advancement in neuroscience theory.
This Research Topic encourages further development of intelligent methods and techniques for image and video processing based on neuroscience theory. This can be achieved by optimizing classical neural networks or building new neural networks by considering some known knowledge, such as designing more realistic neuron models and network hierarchy structures. There are three possible forms of such development: 1) using existing neuroscience theory to further advance methods for processing non-structured signals; 2) gaining inspiration from artificial intelligence systems to guide and deepen research in visual neuroscience; and 3) independently advancing theoretical cognition from both the fields of neuroscience and artificial intelligence.
This Research Topic aims to enhance researchers' understanding of existing artificial intelligence technologies and promote the deep integration and exploration of the development of neuroscience and intelligent visual technology. We hope that neuroscience can guide the design of more powerful biomimetic systems to better serve modern society and humanity. And we also hope that related research brings up promise for advancing our understanding of the brain and developing innovative intelligent technologies. Therefore, we are seeking original research papers on topics including, but not limited to:
Functional artificial neuron model
Brain-like system
Computer-aided diagnosis
Small-data driven learning
Image classification
Image object recognition
Image generation
Video classification
Motion detection
Video object tracking
Video action recognition