MatcherCompass guides image matcher choice for robotics hardware and conditions
MatcherCompass: A Deployment-Aware Benchmark to Guide Image Matcher Selection in the Wild
Robotics
Summary
Robots that explore environments need to match images accurately and quickly, even as conditions like lighting and sensor type change. The paper introduces MatcherCompass, a tool that compares nine image matching methods across different image sizes, computing platforms, and visual conditions. This helps robot builders pick the best image matcher that fits their device’s speed, memory, and energy limits. MatcherCompass provides clear advice by measuring how matcher choice changes depending on hardware and visual challenges.
What this means in practice
- •For field robotics engineers: Select image matching algorithms optimized for onboard hardware and varying visual conditions to improve robot navigation accuracy and efficiency.
- •For autonomous vehicle developers: Choose image matching pipelines adapted to different sensors and computational resources for reliable localization under day and night or thermal conditions.
Authors
Hyunwoo Kim, Giseop Kim
Abstract
Field robots operating across time of day and sensing modalities require accurate image correspondences within onboard time and resource budgets. However, accuracy and runtime reported for individual methods on a single device provide limited guidance for choosing a matcher and its configuration on a target platform. We present MatcherCompass, a deployment-aware benchmark for choosing local feature matchers in field robotics. Under common input and pose-evaluation procedures, we compare nine classical and learned matching pipelines across four image resolutions and supported numerical precisions. Four visual conditions cover viewpoint variation, day--night matching in visible and thermal imagery, and daytime visible--thermal matching. We evaluate pose accuracy using the area under the error--recall curve (AUC) at $5^\circ$, $10^\circ$, and $20^\circ$, and measure runtime, GPU memory, and energy per image pair on four GPU platforms spanning workstation and onboard computers. The results show that changes in hardware, input resolution, and numerical precision can move a matcher across a runtime budget boundary, altering the feasible choices. We organize the measurements into a selection guide that returns all configurations satisfying user-specified time and resource limits, together with their accuracy under the selected visual condition. MatcherCompass provides measured evidence for choosing matching pipelines that fit a robot's sensing conditions and computing hardware. Project page: https://matchercompass.github.io/.