Learning Deep NBNN Representations for Robust Place Categorization

TL;DR

This paper introduces a novel deep learning approach combining CNN features with a Naive Bayes Nearest Neighbor model for robust semantic place categorization, outperforming previous methods especially under occlusions and environmental changes.

Contribution

It integrates NBNN with CNNs into a fully-convolutional network for improved place recognition robustness and accuracy.

Findings

Outperforms previous CNN-based place recognition methods.

Robust to occlusions, environmental, and sensor variations.

Effective in challenging robot place recognition scenarios.

Abstract

This paper presents an approach for semantic place categorization using data obtained from RGB cameras. Previous studies on visual place recognition and classification have shown that, by considering features derived from pre-trained Convolutional Neural Networks (CNNs) in combination with part-based classification models, high recognition accuracy can be achieved, even in presence of occlusions and severe viewpoint changes. Inspired by these works, we propose to exploit local deep representations, representing images as set of regions applying a Na\"{i}ve Bayes Nearest Neighbor (NBNN) model for image classification. As opposed to previous methods where CNNs are merely used as feature extractors, our approach seamlessly integrates the NBNN model into a fully-convolutional neural network. Experimental results show that the proposed algorithm outperforms previous methods based on pre-trained CNN models and that, when employed in challenging robot place recognition tasks, it is robust to occlusions, environmental and sensor changes.