Learning Deep NBNN Representations for Robust Place Categorization

Abstract

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 pretrained 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 Naïve 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.

Development of a Highly Precise Place Recognition Module for Effective Human-robot Interactions in Changing Lighting and Viewpoint Conditions

“…Robust place recognition for mobile robots is one of the major challenges of HRI research. The high complexity of robust place recognition comes from changes in the scene [16], varying lighting and viewpoint conditions [23], the limited computational resources [18] and the high complexity of the places [1].…”

“…These feature descriptors extract textural features such as edges or bright and dark spots, which can be used to categorize the image [15]. Due to the high variability of environments conditions, handcrafted descriptors struggle with robustness [1,17].…”

“…In many homes, domestic robots, self-driving cars and voice-activated assistants and interactions with them are common nowadays. One of the most important aspects of Human-Robot-Interaction (HRI) is the robots ability to understand abstract spatial concepts and act accordingly [1,2]. For example, a robot vacuum cleaner may be asked to clean the bedroom, therefore the robot's recognition of a bedroom should match the human's understanding of such a place [1].…”

“…Our approach is robust against severe changes in lighting and viewpoint conditions [1] and shows good performance in different environments [15].…”

Robust Place Categorization With Deep Domain Generalization

“…due to the presence of people or obstacles and to changing lighting conditions) make this task extremely challenging. Traditional place categorization approaches [7], [8], [9], [10] require labeled datasets of training images. While the resulting models are very accurate when test samples are similar to training data, their performance significantly degrade when the robot collects images with very different visual appearance [11].…”