Scribd
Upload
36 views11 pages

Comparitive Study On Mobile Based Landmark Recognition Techniques

This document compares techniques for mobile-based landmark recognition, noting that it must account for both visual content analysis and location context due to mobile devices' limitations of processing power, storage, and real-time requirements. It analyzes content-based approaches like key points, dense sampling and segmentation, finds that local features like SIFT perform better than global features, and concludes that an enhanced bag-of-words method shows the most promise with the best recognition rate and fast processing speed.

Uploaded by

santhipriyasunkara
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
36 views11 pages

Comparitive Study On Mobile Based Landmark Recognition Techniques

This document compares techniques for mobile-based landmark recognition, noting that it must account for both visual content analysis and location context due to mobile devices' limitations of processing power, storage, and real-time requirements. It analyzes content-based approaches like key points, dense sampling and segmentation, finds that local features like SIFT perform better than global features, and concludes that an enhanced bag-of-words method shows the most promise with the best recognition rate and fast processing speed.

Uploaded by

santhipriyasunkara
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 11

COMPARITIVE STUDY ON MOBILE BASED LANDMARK RECOGNITION TECHNIQUES

S SANTHI PRIYA

Non mobile-based landmark recognition is usually considered a content-based system.


visual features

mobile based landmark recognition should be considered an integrated content- and context based system.
Location

The design of a mobile landmark recognition system must take into account the following inherent challenges and limitations of mobile devices: Low processing power. Limited storage capacity. Real-time requirements.

CONTENT ANALYSIS
Key point based approach Dense sampling approach Segmentation based approach

Context analysis

Conclusion Local features work better than global features. This observation is supported by statistical hypothesis testing. SIFT has superior performance when compared to many other local descriptors. The dense-patch approach offers better performance than the sparse keypoint-based and segmentation based techniques. The enhanced BoW method is the most promising method, with the best recognition rate and fast processing speed.

You might also like