Fully Automated Annotation With Noise-Masked Visual Markers for Deep-Learning-Based Object Detection

Abstract

Training deep-learning-based vision systems requires the manual annotation of a significant amount of data to optimize several parameters of the deep convolutional neural networks. Such manual annotation is highly time-consuming and labor-intensive. To reduce this burden, a previous study presented a fully automated annotation approach that does not require any manual intervention. The proposed method associates a visual marker with an object and captures it in the same image. However, because the previous method relied on moving the object within the capturing range using a fixed-point camera, the collected image dataset was limited in terms of capturing viewpoints. To overcome this limitation, this study presents a mobile application-based free-viewpoint image-capturing method. With the proposed application, users can collect multi-view image datasets automatically that are annotated with bounding boxes by moving the camera. However, capturing images through human involvement is laborious and monotonous. Therefore, we propose gamified application features to track the progress of the collection status. Our experiments demonstrated that using the gamified mobile application for bounding box annotation, with visible collection progress status, can motivate users to collect multi-view object image datasets with less mental workload and time pressure in an enjoyable manner, leading to increased engagement.

Robotic Waste Sorter With Agile Manipulation and Quickly Trainable Detector

“…Fig. 3 shows the proposed dataset collection procedure with its automatic annotation method [19]. Fig.…”

“…To demonstrate the effectivity of the robotic training dataset collection system compared with the collection methods previously proposed in [19], [21], this section describes the results of the comparison of times needed to collect image datasets. Table 1 shows the average time needed to collect 100 images for one target object and the method (automatic or manual) for three processes: object replacement, image acquisition, and annotation.…”

“…The robot arm and rotating stage are automatically controlled while capturing images. Our previously proposed automatic annotation method [19] using augmented-reality (AR) marker detection [20] is applied to captured images. To train the DL-based waste detector, we place the collection-target object on the rotating table for image capture.…”

Identifying Reflected Images From Object Detector in Indoor Environment Utilizing Depth Information

“…I N the past few years, there was a huge advance in object detection attributed to the development of deep learning. Detection algorithms such as Faster R-CNN [1], YOLO [2], SSD [3], and various methodologies [4], [5] enabled object detection to operate robustly and in real-time, so they are being used in many real-world applications. However, when applying these to actual robot systems, many false detections can occur.…”

A Study on Fight Against COVID-19 from Latest Technological Intervention

“…(II) Diagnosis and Treatment Aspect Early Diagnosis and fast treatment will save lives. Diagnosis using deep learning assists radiologist from saving their effort and time to a greater extent and arrives faster conclusion [8,20]. (III) Prevention Aspect Machine learning and Computer vision as an aid for drug discovery and Monitoring and enforce social distancing through visual social distancing [4,15].…”