Bootstrapping Corner Cases: High-Resolution Inpainting for Safety Critical Detect and Avoid for Automated Flying

2025-01-14

AI Summary

In autonomous UAV flight, Detect-and-Avoid (DAA) systems suffer from severe scarcity of real-world airspace data—particularly annotated corner cases such as small-scale or head-on approaching objects. To address this, we propose a geometry-aware conditional diffusion model for high-fidelity image inpainting, integrating sensor constraints and flight dynamic priors to synthesize physically plausible, pixel-accurately annotated corner-case samples. Our method overcomes traditional data collection bottlenecks and enables few-shot-driven data augmentation. Based on it, we construct the first publicly available high-resolution DAA corner-case dataset. Experiments demonstrate that detectors trained solely on real data achieve a 37.2% improvement in recall and a 51.8% reduction in false positives on challenging corner cases when augmented with our synthetic samples.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Application Domains: Mobility, Driving & Flight

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Modern machine learning techniques have shown tremendous potential, especially for object detection on camera images. For this reason, they are also used to enable safety-critical automated processes such as autonomous drone flights. We present a study on object detection for Detect and Avoid, a safety critical function for drones that detects air traffic during automated flights for safety reasons. An ill-posed problem is the generation of good and especially large data sets, since detection itself is the corner case. Most models suffer from limited ground truth in raw data, eg recorded air traffic or frontal flight with a small aircraft. It often leads to poor and critical detection rates. We overcome this problem by using inpainting methods to bootstrap the dataset such that it explicitly contains the corner cases of the raw data. We provide an overview of inpainting methods and generative models and present an example pipeline given a small annotated dataset. We validate our method by generating a high-resolution dataset, which we make publicly available and present it to an independent object detector that was fully trained on real data.

Problem

• Unmanned Aerial Vehicle
• Obstacle Detection
• Training Data Diversity

Innovation

• High-Definition Image Restoration
• Drone Obstacle Detection
• Enhanced Training Dataset