TY - JOUR
T1 - Comparison of local feature extraction paradigms applied to visual SLAM
AU - López-López, Víctor R.
AU - Trujillo, Leonardo
AU - Legrand, Pierrick
AU - Díaz-Ramírez, Victor H.
AU - Olague, Gustavo
N1 - Funding Information:
Provided by CONACYT Basic Science Research Project No. 178323, DGEST (Mexico) Research Projects No. 5149.13-P, 5414.14-P and TIJ-ING-2012-110, and IRSES project ACoBSEC financed by the European Commission. The first author was supported by CONACYT scholarship No. 302532.
PY - 2016
Y1 - 2016
N2 - The detection and description of locally salient regions is one of the most widely used low-level processes in modern computer vision systems. The general approach relies on the detection of stable and invariant image features that can be uniquely characterized using compact descriptors. Many detection and description algorithms have been proposed, most of them derived using different assumptions or problem models. This work presents a comparison of different approaches towards the feature extraction problem, namely: (1) standard computer vision techniques; (2) automatic synthesis techniques based on genetic programming (GP); and (3) a new local descriptor based on composite correlation filtering, proposed for the first time in this paper. The considered methods are evaluated on a difficult real-world problem, vision-based simultaneous localization and mapping (SLAM). Using three experimental scenarios, results indicate that the GP-based methods and the correlation filtering techniques outperform widely used computer vision algorithms such as the Harris and Shi-Tomasi detectors and the Speeded Up Robust Features descriptor.
AB - The detection and description of locally salient regions is one of the most widely used low-level processes in modern computer vision systems. The general approach relies on the detection of stable and invariant image features that can be uniquely characterized using compact descriptors. Many detection and description algorithms have been proposed, most of them derived using different assumptions or problem models. This work presents a comparison of different approaches towards the feature extraction problem, namely: (1) standard computer vision techniques; (2) automatic synthesis techniques based on genetic programming (GP); and (3) a new local descriptor based on composite correlation filtering, proposed for the first time in this paper. The considered methods are evaluated on a difficult real-world problem, vision-based simultaneous localization and mapping (SLAM). Using three experimental scenarios, results indicate that the GP-based methods and the correlation filtering techniques outperform widely used computer vision algorithms such as the Harris and Shi-Tomasi detectors and the Speeded Up Robust Features descriptor.
KW - Composite correlation filter
KW - Genetic programming
KW - Local features
KW - SLAM
UR - http://www.scopus.com/inward/record.url?scp=85007330968&partnerID=8YFLogxK
U2 - 10.13053/CyS-20-4-2500
DO - 10.13053/CyS-20-4-2500
M3 - Artículo
SN - 1405-5546
VL - 20
SP - 565
EP - 587
JO - Computacion y Sistemas
JF - Computacion y Sistemas
IS - 4
ER -