DeepBbox: Accelerating Precise Ground Truth Generation for Autonomous Driving Datasets

TL;DR

DeepBbox is an algorithm that automates the correction of loose object labels into precise bounding boxes, significantly reducing manual annotation effort and increasing labeling accuracy in autonomous driving datasets.

Contribution

It introduces a novel method for automatic bounding box correction that improves annotation efficiency and accuracy in large-scale autonomous driving datasets.

Findings

Increases automatic labeling of object edges by 50% within 1% error

Reduces manual annotation time significantly

Improves bounding box precision in dataset annotations

Abstract

Autonomous driving requires various computer vision algorithms, such as object detection and tracking.Precisely-labeled datasets (i.e., objects are fully contained in bounding boxes with only a few extra pixels) are preferred for training such algorithms, so that the algorithms can detect exact locations of the objects. However, it is very time-consuming and hence expensive to generate precise labels for image sequences at scale. In this paper, we propose DeepBbox, an algorithm that corrects loose object labels into right bounding boxes to reduce human annotation efforts. We use Cityscapes dataset to show annotation efficiency and accuracy improvement using DeepBbox. Experimental results show that, with DeepBbox,we can increase the number of object edges that are labeled automatically (within 1\% error) by 50% to reduce manual annotation time.