The project is centered on utilizing the principles of Deep Learning and Computer Vision to accurately classify objects in Images. The idea behind this objective is to utilize the same principles in an Advanced Manufacturing setting. We have built our own Architecture (completely different from the resources in the paperspace blog) to effectively train the RCNN model. We have achieved this in the pre-processing step by segmenting the predefined classes in training data by manually annotating them in LabelMe (which is more effective in generating JSON files than LabelImg). This dataset is used to train the RCNN and improve it further with help of taking epoch losses as a reference.
Object Detection in Advanced Manufacturing using RCNN
The built model utilizes principles of Deep Learning and Computer vision to detect and classify objects in illustrations.
Updates
Leave feedback in the comments!
Log in or sign up for Devpost to join the conversation.