Towards training-free appearance-based localization: Probabilistic models for whole-image descriptors

Abstract

Whole image descriptors have been shown to be remarkably robust to perceptual change especially compared to local features. However, whole-image-based localization systems typically rely on heuristic methods for determining appropriate matching thresholds in a particular environment. These environment-specific tuning requirements and the lack of a meaningful interpretation of arbitrary thresholds limit the general applicability of these systems. In this paper we present a Bayesian model of probability for whole-image descriptors that can be seamlessly integrated into localization systems designed for probabilistic visual input. We demonstrate this method using CAT-Graph, an appearance-based visual localization system originally designed for a FAB-MAP-style probabilistic input. We show that using whole-image descriptors as visual input extends CAT-Graph's functionality to environments that experience a greater amount of perceptual change. We also present a method of estimating whole-image probability models in an online manner, removing the need for a prior training phase. We show that this online, automated training method can perform comparably to pretrained, manually tuned local descriptor methods.

Place Recognition with ConvNet Landmarks: Viewpoint-Robust, Condition-Robust, Training-Free

“…Place Recognition: Methods that address the place recognition problem span from matching sequences of images [27,17,40,33,29], transforming images to become invariant against common scene changes such as shadows [6,43,25,24,21], learning how environments change over time and predicting these changes in image space [30,21,31], particle filterbased approaches that build up place recognition hypotheses over time [23,39,22], or build a map of experiences that cover the different appearances of a place over time [5].…”

On the performance of ConvNet features for place recognition

“…Visual place recognition -the ability to recognize a known place in the environment using vision as the main sensor modality -is largely affected by these appearance changes and is therefore an active research field within the robotics community. The recent literature proposes a variety of approaches to address the challenges of this field [1]- [14]. Recent progress in the computer vision and machine learning community has shown that the features generated by Convolutional Networks (Con-vNets, see Fig.…”

Automatic Coverage Selection for Surface-Based Visual Localization

“…Due to these difficulties, there have been several research areas investigating the development of automatic calibration routines to improve the performance of visual localization alogrithms. Lowry et al demonstrated online training-free procedures that could determine the probabilistic model for evaluating whether a query image came from the same location as a reference image, even under significant appearance variation [22], [23]. In [24], [25] and [26] Jacobson et al explored novel calibration methods to automatically optimize sensor threshold parameters for place recognition.…”