Abstract
Highly accurate, camera-based object detection is an essential component of autonomous navigation and assistive technologies. In particular, for on-road applications, localization quality of objects in the image plane is important for accurate distance estimation, safe trajectory prediction, and motion planning. In this paper, we mathematically formulate and study a strategy for improving object localization with a deep convolutional neural network. An iterative region-of-interest pooling framework is proposed for predicting increasingly tight object boxes and addressing limitations in current state-of-the-art deep detection models. The method is shown to significantly improve the performance on a variety of datasets, scene settings, and camera perspectives, producing high-quality object boxes at a minor additional computational expense. Specifically, the architecture achieves impressive gains in performance (up to 6% improvement in detection accuracy) at fast run-time speed (0.22 s per frame on $1242 \times 375$ sized images). The iterative refinement is shown to impact subsequent vision tasks, such as object tracking in the image plane and in ground plane.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have