BinVPR: Binary Neural Networks towards Real-Valued for Visual Place Recognition

• Published in: Sensors (Basel, Switzerland) Vol. 24; no. 13; p. 4130
• Main Authors: Wang, Junshuai, Han, Junyu, Dong, Ruifang, Kan, Jiangming
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 25.06.2024
• Subjects: Accuracy; binary neural networks; Computer vision; Datasets; Deep learning; Efficiency; Energy consumption; gradient mismatch; gradient vanishing; Hypotheses; Image retrieval; Investigations; Localization; model compression; Neural networks; Principles; Robots; visual place recognition; model compression; binary neural networks; gradient mismatch; gradient vanishing; visual place recognition

Visual Place Recognition (VPR) aims to determine whether a robot or visual navigation system locates in a previously visited place using visual information. It is an essential technology and challenging problem in computer vision and robotic communities. Recently, numerous works have demonstrated th...