Edged USLAM: Edge-Aware Event-Based SLAM with Learning-Based Depth Priors

TL;DR

Edged USLAM is presented, a hybrid visual-inertial system that extends Ultimate SLAM (USLAM) with an edge-aware front-end and a lightweight depth module that provides superior stability and minimal drift in slow or structured trajectories, ensuring consistently accurate localization on real flights under challenging illumination.

Abstract

Conventional visual simultaneous localization and mapping (SLAM) algorithms often fail under rapid motion, low illumination, or abrupt lighting transitions due to motion blur and limited dynamic range. Event cameras mitigate these issues with high temporal resolution and high dynamic range (HDR), but their sparse, asynchronous outputs complicate feature extraction and integration with other sensors; e.g. inertial measurement units (IMUs) and standard cameras. We present Edged USLAM, a hybrid visual-inertial system that extends Ultimate SLAM (USLAM) with an edge-aware front-end and a lightweight depth module. The frontend enhances event frames for robust feature tracking and nonlinear motion compensation, while the depth module provides coarse, region-of-interest (ROI)-based scene depth to improve motion compensation and scale consistency. Evaluations across public benchmarks and real-world unmanned air vehicle (UAV) flights demonstrate that performance varies significantly by scenario. For instance, event-only methods like point-line event-based visual-inertial odometry (PL-EVIO) or learning-based pipelines such as deep event-based visual odometry (DEVO) excel in highly aggressive or extreme HDR conditions. In contrast, Edged USLAM provides superior stability and minimal drift in slow or structured trajectories, ensuring consistently accurate localization on real flights under challenging illumination. These findings highlight the complementary strengths of event-only, learning-based, and hybrid approaches, while positioning Edged USLAM as a robust solution for diverse aerial navigation tasks.

Figures (7)

• Figure 1: Overview of the proposed Edged USLAM framework. The UAV (right) equipped with a protective cage, event camera, and depth module navigates through a cluttered indoor environment (top). The pipeline integrates learning-based depth estimation (bottom left) with enhanced event frames for robust ORB-based tracking (bottom center-right), enabling accurate 3D trajectory reconstruction (top).
• Figure 2: Overview of the proposed Edged USLAM methodology. Events, standard frames, and IMU data are synchronized and fused with nonlinear motion compensation, followed by frame enhancement and ORB-based feature tracking. Landmarks are refined via RANSAC and triangulation, while depth priors from the ResNet module support consistent state estimation and optimization in the back-end.
• Figure 3: Comparison of event frame enhancement methods. (a) Motion compensation sharpens asynchronous event frames into a more stable representation. (b) CLAHE improves local contrast under varying illumination. (c) Canny edge detection extracts fine-grained structural contours. (d) Laplacian filtering highlights high-frequency edges and textures.
• Figure 4: Real-time demonstration of event-based depth estimation. Top row: sample event inputs (a–b). Bottom row: corresponding coarse depth estimation results (c–d) computed by a lightweight model for rapid inference, intended not for accurate reconstruction but for rapid inference of approximate object distances.
• Figure 5: TESTUDO an UAV platform equipped with a collision-tolerant protective cage and multi-sensor payload (event camera, tracking sensor, and IMU) for real-time SLAM evaluation in confined and hazardous environments