Efficiently Collecting Training Dataset for 2D Object Detection by Online Visual Feedback

TL;DR

A human-in-the-loop dataset-collection method using a web application to counterbalance workload and performance by encouraging the collection of multi-view object image datasets enjoyably, thereby amplifying motivation.

Abstract

Training deep-learning-based vision systems require the manual annotation of a significant number of images. Such manual annotation is highly time-consuming and labor-intensive. Although previous studies have attempted to eliminate the effort required for annotation, the effort required for image collection was retained. To address this, we propose a human-in-the-loop dataset collection method that uses a web application. To counterbalance the workload and performance by encouraging the collection of multi-view object image datasets in an enjoyable manner, thereby amplifying motivation, we propose three types of online visual feedback features to track the progress of the collection status. Our experiments thoroughly investigated the impact of each feature on collection performance and quality of operation. The results suggested the feasibility of annotation and object detection.

Figures (19)

• Figure 1: Display transitions of the developed mobile application.
• Figure 2: System configuration for mobile application.
• Figure 3: Annotation flow
• Figure 4: Superimposed annotation
• Figure 6: Visualization of collected datasets.