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

2026-04-07

AI Summary

This work addresses the challenge of detecting ultra-long-range objects (>500 meters) in high-speed autonomous driving, where such targets often span only a few pixels and thus evade detection by existing models. To overcome this limitation, the authors propose a two-stage detection framework featuring a learnable hyperbolic foveal resampling layer that integrates high-resolution image processing with multi-scale feature fusion. This design effectively enhances the representational capacity for minute objects without substantially increasing computational overhead. Experimental results demonstrate that the proposed method improves the mean average precision (mAP) for detections beyond 250 meters from 0.185 to 0.326—a relative gain of 76%—while maintaining robust performance across all distance 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.

Problem

• ultra-long-range object detection
• autonomous driving
• small object detection
• image-based perception
• sensor limitations

Innovation

• learnable hyperbolic foveation
• ultra-long-range object detection
• resampling layer
• autonomous driving
• small object detection