Avoiding to face the challenges of visual place recognition
Autor: | Ehsan Mihankhah, Danwei Wang |
---|---|
Přispěvatelé: | School of Electrical and Electronic Engineering, Intelligent Systems Conference 2018 (IntelliSys 2018) |
Jazyk: | angličtina |
Rok vydání: | 2018 |
Předmět: |
0209 industrial biotechnology
Computer science Scale (chemistry) ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Cognitive neuroscience of visual object recognition 02 engineering and technology Place Detection Place Recognition 020901 industrial engineering & automation Human–computer interaction Face (geometry) 0202 electrical engineering electronic engineering information engineering Selection (linguistics) Electrical and electronic engineering [Engineering] Table (database) 020201 artificial intelligence & image processing Recognition algorithm |
Zdroj: | Advances in Intelligent Systems and Computing ISBN: 9783030010539 IntelliSys (1) |
Popis: | Through this paper, bottlenecks of conventional place recognition techniques are studied, and a replacement strategy is proposed for each item. Conventional place recognition algorithms are extensions of object recognition techniques applied to larger scale targets known as the place landmarks. The discussion presented in this paper aims to address the challenges of detection and recognition of the places, which make this topic distinctive from detection and recognition of the objects and landmarks. The challenges are listed under related categories. The table of challenges, reasons, and the recommendations to avoid these situations is presented as the guideline for selection of proper tools for place recognition purpose. Accepted version |
Databáze: | OpenAIRE |
Externí odkaz: |