Content-Based Representation and Retrieval of Visual Media: A State-of-the-Art Review
read more
Citations
Content-based image retrieval at the end of the early years
Image retrieval: Ideas, influences, and trends of the new age
Bursty and Hierarchical Structure in Streams
Bursty and hierarchical structure in streams
Similarity measures
References
Query by image and video content: the QBIC system
Similarity of color images
Photobook: content-based manipulation of image databases
Automatic partitioning of full-motion video
Photobook: tools for content-based manipulation of image databases
Related Papers (5)
Frequently Asked Questions (15)
Q2. What is the main reason for content based retrieval?
Not only content based retrieval reduces the high variability among human indexers, but it also enables more “fuzzy” browsing and search which in many application is an essential part of the process.
Q3. What are the main applications of multimedia technology?
The present development of multimedia technology and information highways has put content processing of visual media at the core of key application domains: digital and interactive video, large distributed digital libraries, multimedia publishing.
Q4. What is the problem of combining color similarity with a spatial component?
One gets three components (periodic, evanescent and random) corresponding to the bi-dimensional periodicity, mono-dimensional orientation, and complexity of the analyzed texture.
Q5. What types of similarity can be considered in a given image?
Gudivada has listed possible types of similarity for retrieval in[Gudivada95]: color similarity, texture similarity, shape similarity, spatial similarity, etc.
Q6. Why is color distribution similarity one of the first choices?
Color distribution similarity has been one of the first choices [HK92, FSN+95] because if one chooses a proper representation and measure it can be partially reliable even in presence of changes in lighting, view angle, and scale.
Q7. what is the method to locate images with a particular face?
When the problem is to locate images with a particular object (a particular face, a particular building) and not any object of a given type, principal component analysis methods of more general features of the images is the only efficient method.
Q8. What are the problems of the present schemes for shape similarity modelling?
As for color and texture, the present schemes for shape similarity modelling are faced with serious difficulties when images include several objects or background.
Q9. What is the main reason for using color distributions in search?
It has been proposed in [AJL95] to use hue and saturation distributions only when one wants to capture lighting-independent color distribution properties which are good signatures of a scene when the scale does not change too much.
Q10. What is the time structure for retrieval of visual media?
Content-based retrieval of visual media and representation of visual documents in human-computer interfaces are based on the availability of content representation data (time-structure for1.
Q11. What are the main reasons for the need for content processing techniques?
The need for content processing techniques has been made evident from a variety of angles, ranging from achieving better quality in compression, allowing user choice of programs in video-on-demand, achieving better productivity in video production, providing access to large still image databases or integrating still images and video in multimedia publishing and cooperative work.
Q12. What is the problem of combining texture information with a spatial component?
in this case, the researchers have tried to define shape representations which are scale independent, resting on curvature, angle statistics and contour complexity.
Q13. What is the definition of shape similarity?
A proper definition of shape similarity calls for the distinctions between shape similarity in5 images (similarity between actual geometrical shapes appearing in the images) and shape similarity between the objects depicted by the images, i.e. similarity modulo a number of geometrical transformations corresponding to changes in view angle, optical parameters and scale.
Q14. When did researchers start looking at retrieval by similarity in large set of heterogeneous?
Apart from paper models [Aig87], it was only in the beginning of the 90s that researchers started to look at retrieval by similarity in large set of heterogeneous images with no specific model of their semantic contents.
Q15. What are examples of applications where visual appearance is important?
Examples of such applications are where visual appearance (e.g. color, texture, shape, motion) are important search arguments like in stock photo/video, art, retail, on-line shopping etc.