Interdisciplinary studies involving the visual system, computational models and computer vison, show great potential in promoting the joint development of these fields. The achievements and knowledge obtained through such research assists not only in understanding of the sensory system as a biological system, but can also inspire novel experimental studies and development of new computational models. Furthermore, these studies can provide new ideas and approaches for engineering algorithms and applied technologies.
Both the biological visual system and the computer vison system strive to understand and interact in the visual physical world based on noisy, occluded, illuminated, partial and often inconsistent information extracted from the world. Consider even the most basic example of deciding whether 2 collinear line segments originate from two independent entities in the world or in fact originate from a single continuous line source in the world that is partially occluded. Both biological and computer systems must make educated "guesses" in order to interpret such inputs. The notion of educated guesses is used at all levels of the visual system in order to interpret the world ranging from contour and shape estimation, color and texture analysis, 3D structure understanding, continuous motion interpretation and up to action understanding and more. Studies of the human visual system attempt to understand how the biological system performs these educated guesses, whereas the computer vision systems attempt to develop algorithms and technology to automatically make these educated guesses. Mathematical, computational and physical models are often used to model the biological system on one hand and to guide the algorithms and technologies on the other.
The proposed topic for the Research Topic focuses on 2 types of studies: 1) on visual system mechanisms and their computational models, which in turn inspire or have the potential to inspire applicative algorithms and 2) on mathematical and physical models of the world used for implementing algorithms that can predict human visual perception and behavior. (For this criteria functional minimization models and machine learning models can be accepted.)
The contributions will cover various aspects of the visual system ranging from the photoreceptors in the retina, to early and late visual areas processing and further, into perceptual phenomena modeling including illusions, attentional effects, and higher level tasks such as object classification, activity detection and more.
Articles in the following areas are of particular relevance to this research topic:
1) Receptive field models of cells in different visual areas
2) Feed-forward and feedback models of the ventral and dorsal pathways of the visual cortex.
3) Adaptation mechanisms of the different visual areas and pathways.
4) Scale-, occlusion-, orientation- invariant modeling of the visual cortex.
5) Visual adaptation mechanisms underlying perception phenomena, such as induction in the chromatic, spatial and temporal domains (preferably demonstrated or applied also on images )
6) Encoding and decoding visual perception using functional or machine learning techniques.
7) Depth and motion perception models.
8) Color vision models that relate to sensors, color coding receptive fields and chromatic effects.
9) Roles of attention and derived algorithms and technologies.
10) Contribution of visual models to real images such as for enhancing medical images.
Interdisciplinary studies involving the visual system, computational models and computer vison, show great potential in promoting the joint development of these fields. The achievements and knowledge obtained through such research assists not only in understanding of the sensory system as a biological system, but can also inspire novel experimental studies and development of new computational models. Furthermore, these studies can provide new ideas and approaches for engineering algorithms and applied technologies.
Both the biological visual system and the computer vison system strive to understand and interact in the visual physical world based on noisy, occluded, illuminated, partial and often inconsistent information extracted from the world. Consider even the most basic example of deciding whether 2 collinear line segments originate from two independent entities in the world or in fact originate from a single continuous line source in the world that is partially occluded. Both biological and computer systems must make educated "guesses" in order to interpret such inputs. The notion of educated guesses is used at all levels of the visual system in order to interpret the world ranging from contour and shape estimation, color and texture analysis, 3D structure understanding, continuous motion interpretation and up to action understanding and more. Studies of the human visual system attempt to understand how the biological system performs these educated guesses, whereas the computer vision systems attempt to develop algorithms and technology to automatically make these educated guesses. Mathematical, computational and physical models are often used to model the biological system on one hand and to guide the algorithms and technologies on the other.
The proposed topic for the Research Topic focuses on 2 types of studies: 1) on visual system mechanisms and their computational models, which in turn inspire or have the potential to inspire applicative algorithms and 2) on mathematical and physical models of the world used for implementing algorithms that can predict human visual perception and behavior. (For this criteria functional minimization models and machine learning models can be accepted.)
The contributions will cover various aspects of the visual system ranging from the photoreceptors in the retina, to early and late visual areas processing and further, into perceptual phenomena modeling including illusions, attentional effects, and higher level tasks such as object classification, activity detection and more.
Articles in the following areas are of particular relevance to this research topic:
1) Receptive field models of cells in different visual areas
2) Feed-forward and feedback models of the ventral and dorsal pathways of the visual cortex.
3) Adaptation mechanisms of the different visual areas and pathways.
4) Scale-, occlusion-, orientation- invariant modeling of the visual cortex.
5) Visual adaptation mechanisms underlying perception phenomena, such as induction in the chromatic, spatial and temporal domains (preferably demonstrated or applied also on images )
6) Encoding and decoding visual perception using functional or machine learning techniques.
7) Depth and motion perception models.
8) Color vision models that relate to sensors, color coding receptive fields and chromatic effects.
9) Roles of attention and derived algorithms and technologies.
10) Contribution of visual models to real images such as for enhancing medical images.
