Most of the current state-of-the-art frameworks for cross- season visual place recognition (CS-VPR) focus on domain adaptation (DA) to a single specific season. From the viewpoint of long-term CS-VPR, such frameworks do not scale well to sequential multiple domains (e.g., spring → summer → autumun → winter → ⋯). The goal of this study is to develop a novel long-term ensemble learning (LEL) framework that allows for a constant cost retraining in long- term sequential-multi-domain CS-VPR (SMD-VPR), which only requires the memorization of a small constant number of deep convolutional neural networks (CNNs) and can retrain the CNN ensemble of every season at a small constant time/space cost. We frame our task as the multi-teacher multi-student knowledge distillation (MTMS-KD), which re- cursively compresses all the previous season’s knowledge into a current CNN ensemble. We further address the issue of teacher-student-assignment (TSA) to achieve a good general- ization/specialization tradeoff. Experimental results on SMD- VPR tasks validate the efficacy of the proposed approach.
Long-Term Knowledge Distillation of Visual Place Classifiers
2019-10-01
2121143 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Learning and Calibrating Per-Location Classifiers for Visual Place Recognition
British Library Online Contents | 2016
|Neural classifiers for automated visual inspection
Online Contents | 1994
|Neural Classifiers for Automated Visual Inspection
SAGE Publications | 1994
|Neural classifiers for automated visual inspection
Kraftfahrwesen | 1994
|