Exploring Object-Aware Attention Guided Frame Association for RGB-D SLAM

TL;DR

This paper introduces a novel RGB-D SLAM method that leverages gradient-based, object-aware attention within CNN features to enhance frame association, especially in large indoor environments.

Contribution

It presents a new approach integrating layer-wise attention from network gradients into CNN features for improved SLAM performance.

Findings

Enhanced frame association accuracy in large environments

Improved SLAM robustness through attention-guided features

Superior performance over baseline methods

Abstract

Attention models have recently emerged as a powerful approach, demonstrating significant progress in various fields. Visualization techniques, such as class activation mapping, provide visual insights into the reasoning of convolutional neural networks (CNNs). Using network gradients, it is possible to identify regions where the network pays attention during image recognition tasks. Furthermore, these gradients can be combined with CNN features to localize more generalizable, task-specific attentive (salient) regions within scenes. However, explicit use of this gradient-based attention information integrated directly into CNN representations for semantic object understanding remains limited. Such integration is particularly beneficial for visual tasks like simultaneous localization and mapping (SLAM), where CNN representations enriched with spatially attentive object locations can enhance performance. In this work, we propose utilizing task-specific network attention for RGB-D indoor SLAM. Specifically, we integrate layer-wise attention information derived from network gradients with CNN feature representations to improve frame association performance. Experimental results indicate improved performance compared to baseline methods, particularly for large environments.