Constellation Dataset: Benchmarking High-Altitude Object Detection for an Urban Intersection

TL;DR

The Constellation dataset provides a comprehensive benchmark for high-altitude object detection in urban scenes, highlighting challenges in small pedestrian detection and demonstrating improvements through domain-specific data and pseudo-labeling.

Contribution

This paper introduces the Constellation dataset for high-altitude urban object detection and evaluates state-of-the-art models, revealing performance gaps and benefits of domain-specific data augmentation and pseudo-labeling.

Findings

State-of-the-art models perform 10% worse on small pedestrians compared to vehicles.

Pretraining with structurally similar datasets increases mean AP by 1.8%.

Pseudo-labeling improves model performance, and temporal changes cause performance drift.

Abstract

We introduce Constellation, a dataset of 13K images suitable for research on detection of objects in dense urban streetscapes observed from high-elevation cameras, collected for a variety of temporal conditions. The dataset addresses the need for curated data to explore problems in small object detection exemplified by the limited pixel footprint of pedestrians observed tens of meters from above. It enables the testing of object detection models for variations in lighting, building shadows, weather, and scene dynamics. We evaluate contemporary object detection architectures on the dataset, observing that state-of-the-art methods have lower performance in detecting small pedestrians compared to vehicles, corresponding to a 10% difference in average precision (AP). Using structurally similar datasets for pretraining the models results in an increase of 1.8% mean AP (mAP). We further find that incorporating domain-specific data augmentations helps improve model performance. Using pseudo-labeled data, obtained from inference outcomes of the best-performing models, improves the performance of the models. Finally, comparing the models trained using the data collected in two different time intervals, we find a performance drift in models due to the changes in intersection conditions over time. The best-performing model achieves a pedestrian AP of 92.0% with 11.5 ms inference time on NVIDIA A100 GPUs, and an mAP of 95.4%.