2013
DOI: 10.1007/s11042-013-1503-z
|View full text |Cite
|
Sign up to set email alerts
|

A short-term learning approach based on similarity refinement in content-based image retrieval

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
2
0

Year Published

2013
2013
2022
2022

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 10 publications
(4 citation statements)
references
References 32 publications
0
2
0
Order By: Relevance
“…10-15% improvement as compared to others. Shamsi presents a paper in which they use the RFB approach for modernizing both the masses for a feature type and the masses for feature components [23]. They compare the result with Rui STL and CC STL methods then their result is effective.…”
Section: Related Workmentioning
confidence: 99%
“…10-15% improvement as compared to others. Shamsi presents a paper in which they use the RFB approach for modernizing both the masses for a feature type and the masses for feature components [23]. They compare the result with Rui STL and CC STL methods then their result is effective.…”
Section: Related Workmentioning
confidence: 99%
“…Shape and sketch based descriptors generally require edge detection and image segmentation which limits its applicability. Some other retrieval approaches are quadtree classified vector quantization [3], spatial relations [14], similarity refinement [29] and short and long term learning fusion [23]. Low-level feature descriptions represent the information in the image efficiently and used in several image matching problems but these features are sensitive to the illumination differences (i.e.…”
mentioning
confidence: 99%
“…Kernel Combination [232], Similarity function optimization [30,59], Features and components weighs adjustment [176].…”
Section: Metric Learningmentioning
confidence: 99%
“…Alternatively, with a hybrid approach, Shamsi et al (2014) [176] proposed not only adjusting the different feature weights, but also the weights of each component of the features. The weights of the feature components were adjusted according to the mean and standard deviation values of the features of relevant samples from feedback, while the weight for each feature was adjusted according to the rank positions of the relevant samples on feature specific ranked lists.…”
Section: Metric Learningmentioning
confidence: 99%