Telescope: Learnable Hyperbolic Foveation for Ultra-Long-Range Object Detection

TL;DR

Telescope is a novel two-stage detection model that significantly improves ultra-long-range object detection for autonomous driving by using learnable hyperbolic foveation and image transformation techniques.

Contribution

The paper introduces Telescope, a new model with a re-sampling layer and image transformation that enhances detection of distant objects beyond 250 meters.

Findings

Achieves 76% relative improvement in mAP over state-of-the-art methods at distances beyond 250 meters.

Improves absolute mAP from 0.185 to 0.326 at ultra-long ranges.

Maintains computational efficiency and strong performance across all detection ranges.

Abstract

Autonomous highway driving, especially for long-haul heavy trucks, requires detecting objects at long ranges beyond 500 meters to satisfy braking distance requirements at high speeds. At long distances, vehicles and other critical objects occupy only a few pixels in high-resolution images, causing state-of-the-art object detectors to fail. This challenge is compounded by the limited effective range of commercially available LiDAR sensors, which fall short of ultra-long range thresholds because of quadratic loss of resolution with distance, making image-based detection the most practically scalable solution given commercially available sensor constraints. We introduce Telescope, a two-stage detection model designed for ultra-long range autonomous driving. Alongside a powerful detection backbone, this model contains a novel re-sampling layer and image transformation to address the fundamental challenges of detecting small, distant objects. Telescope achieves $76\%$ relative improvement in mAP in ultra-long range detection compared to state-of-the-art methods (improving from an absolute mAP of 0.185 to 0.326 at distances beyond 250 meters), requires minimal computational overhead, and maintains strong performance across all detection ranges.