About

Pallavi Mitra is a skilled Computer Vision Engineer with 3.94 years of relevant experience in the field. She holds a Master's degree in Computer Science from the University of Minnesota, where she focused on Monocular Depth Estimation using Adversarial Training for her thesis. Pallavi's interests lie in the areas of Computer Vision and Deep Learning, with a particular focus on Multi-View Geometry and Stereo Vision. Pallavi has worked for several reputable companies, including Samsung R&D India, PreNav Inc., and Sentera. At Sentera, she was responsible for designing vision-based algorithms for weed detection and crop counting in the domain of precision agriculture. Pallavi used Active Learning to intelligently select images required for annotation, reducing annotation costs and achieved 75% recall by labeling only 33% of data for detecting individual and patches of weeds in field imagery. She also developed a pipeline to quantify the amount of trust in annotations by a third-party along with an added mechanism to correct poor quality annotations, which helped improve annotation quality by 15%, making a direct impact on the accuracy of deep neural networks. Pallavi achieved an accuracy of 92% in identifying different weed species present in imagery using an Object Classification Network, despite the extremely small sizes of weeds, making identification difficult. Pallavi's skills include software development using Python, C/C++, MATLAB, Tensorflow, Pytorch, and other knowledge and interests such as Machine Learning, Deep Learning, GAN, Non-Linear Optimization, Algorithms, and Virtual Reality. She has also worked as a Graduate Student Researcher at Interactive Robotics and Vision Lab at the University of Minnesota, where she improved Monocular Depth Estimation using Generative Adversarial Networks. Currently, Pallavi is working as a Computer Vision Engineer at Path Robotics, where she is using 3D Computer Vision to build autonomous welding robots. Pallavi's tech stack includes Computer Vision, Deep Learning, GAN, Image Segmentation, Neural Networks, ML, Tensorflow, Pytorch, Software Engineer, Python, C, Research Scientist, and Adversarial Networks.