I am a computer science graduate student at Arizona State University with 2.5 years of professional experience working as a software engineer at Samsung Research Institute, India.
● Evaluated the efficiency of Blitznet architecture for object detection and segmentation in low resolution drone images and analyzed the time vs accuracy tradeoff for various CNN feature extractors.
● Remodeled the Blitznet architecture by replacing the Resnet-50 CNN feature extractor with Inception Resnet v2 by programming in Python and utilizing the Tensorflow library.
● Improved the accuracy for object detection and segmentation in the low resolution drone image dataset by 4%