Research introduces a brand new improvement in landmark retrieval fashions

December 18, 2024

Editors' notes

This text has been reviewed based on Science X's editorial course of and insurance policies. Editors have highlighted the next attributes whereas making certain the content material's credibility:

fact-checked

trusted supply

proofread

Research introduces a brand new improvement in landmark retrieval fashions

A brand new method to landmark retrieval, an space of pc imaginative and prescient that identifies and matches landmark pictures inside a database, is mentioned within the Worldwide Journal of Data and Communication Expertise. The brand new method taken by Kun Tong and GuoXin Tan of the Nationwide Analysis Middle of Cultural Industries at Central China Regular College in Wuhan, improves accuracy and effectivity of picture retrieval techniques and will assist builders navigate advances in pc imaginative and prescient functions akin to object recognition, augmented actuality, and autonomous automobile management.

Landmark retrieval fashions often depend on characteristic descriptors to investigate and examine pictures. These descriptors are available two types: world and native. World descriptors seize the general construction and summary qualities of a picture, whereas native descriptors dwelling in on wonderful particulars akin to textures and spatial preparations. This mix provides complementary details about the picture being analyzed.

Nevertheless, there may be quite a lot of redundancy, which dilutes vital info, resulting in inefficient processing. Furthermore, the fact of captured pictures means variations in viewing angle, lighting circumstances, and the presence of obstructions all result in inaccuracies.

The brand new mannequin makes use of a texture enhancement module to emphasise the essential textural options even in complicated scenes. The module reconstructs characteristic maps to amplify surface-level patterns, making certain that even delicate or distorted textures are highlighted. This may overcome issues that come up due to the viewing angle or poor lighting. The mannequin additionally makes use of a characteristic fusion module that integrates the worldwide and native descriptors to remove redundancies within the knowledge. By prioritizing related particulars and discarding superfluous info, the mannequin streamlines the evaluation to enhance computational effectivity.

Tong and Tan have carried out intensive checks on benchmark datasets, together with the Revisited Oxford and Paris datasets, and present their method to be very efficient and environment friendly at figuring out landmarks.

Extra info: Kun Tong et al, Single-stage landmark retrieval with texture characteristic fusion, Worldwide Journal of Data and Communication Expertise (2024). DOI: 10.1504/IJICT.2024.143328

Supplied by Inderscience Quotation: Research introduces a brand new improvement in landmark retrieval fashions (2024, December 18) retrieved 18 December 2024 from https://techxplore.com/information/2024-12-landmark.html This doc is topic to copyright. Other than any honest dealing for the aim of personal examine or analysis, no half could also be reproduced with out the written permission. The content material is offered for info functions solely.

Discover additional

Wonderful-structure delicate deep studying framework for prediction of catalytic properties with excessive precision shares