Detecting and Grouping Identical Objects for Region Proposal and Classification
This work addresses object detection and classification challenges in scenarios like warehouses, offering incremental improvements in efficiency and accuracy for domain-specific applications.
The paper tackles the problem of detecting and grouping multiple identical object instances in a scene to improve region proposal and classification efficiency. It uses an unsupervised multi-instance object discovery algorithm to provide proposals to a CNN classifier, resulting in fewer regions to evaluate and improved classification accuracy through joint probability modeling.
Often multiple instances of an object occur in the same scene, for example in a warehouse. Unsupervised multi-instance object discovery algorithms are able to detect and identify such objects. We use such an algorithm to provide object proposals to a convolutional neural network (CNN) based classifier. This results in fewer regions to evaluate, compared to traditional region proposal algorithms. Additionally, it enables using the joint probability of multiple instances of an object, resulting in improved classification accuracy. The proposed technique can also split a single class into multiple sub-classes corresponding to the different object types, enabling hierarchical classification.