Publication Date
Spring 2018
Degree Type
Master's Project
Degree Name
Master of Science (MS)
Department
Computer Science
Abstract
Object detection plays a vital role in many real-world computer vision applications such as selfdriving cars, human-less stores and general purpose robotic systems. Convolutional Neural Network(CNN) based Deep Learning has evolved to become the backbone of most computer vision algorithms, including object detection. Most of the research has focused on detecting objects that differ significantly e.g. a car, a person, and a bird. Achieving fine-grained object detection to detect different types within one class of objects from general object detection can be the next step. Fine-grained object detection is crucial to tasks like automated retail checkout. This research has developed deep learning models to detect 200 types of birds of similar size and shape. The models were trained and tested on CUB-200-2011 dataset. To the best of our knowledge, by attaining a mean Average Precision (mAP) of 71.5% we achieved an improvement of 5 percentage points over the previous best mAP of 66.2%.
Recommended Citation
Dalal, Rahul, "FINE-GRAINED OBJECT DETECTION" (2018). Master's Projects. 609.
DOI: https://doi.org/10.31979/etd.f9gs-dd3b
https://scholarworks.sjsu.edu/etd_projects/609
Comments
Object detection plays a vital role in many real-world computer vision applications such as selfdriving cars, human-less stores and general purpose robotic systems. Convolutional Neural Network(CNN) based Deep Learning has evolved to become the backbone of most computer vision algorithms, including object detection. Most of the research has focused on detecting objects that differ significantly e.g. a car, a person, and a bird. Achieving fine-grained object detection to detect different types within one class of objects from general object detection can be the next step. Fine-grained object detection is crucial to tasks like automated retail checkout. This research has developed deep learning models to detect 200 types of birds of similar size and shape. The models were trained and tested on CUB-200-2011 dataset. To the best of our knowledge, by attaining a mean Average Precision (mAP) of 71.5% we achieved an improvement of 5 percentage points over the previous best mAP of 66.2%.