Shivaank Agarwal Email & Phone Number

Sunnyvale, California, United States

Working on deep learning/computer vision algorithms for the Apple Vision Pro

San Diego, California, United States

Fall '21: TA for MATH 11 - Calculus-Based Introductory Probability and Statistics Winter '22: TA for MATH 183 - Statistical MethodsSpring '22: TA for MATH 183 - Statistical MethodsFall '22: TA for CSE 252A - Computer Vision IWinter '23: TA for CSE 252B - Computer Vision IISpring' 23: TA for DSGN1 - Design of everyday things

Sunnyvale, California, United States

Neural architecture search in computer vision domain

Bengaluru, Karnataka, India

1. Improved and optimized sound classification for real-time audio event detection, which performed better than the state-of-the-art YamNet (deep neural network) model on in-domain sounds.2. Collaborated with IoT team to create a video summarization algorithm in compressed h264 domain forsurveillance cameras, resulting in greater performance than comparable compressed domain approaches.

Mumbai, Maharashtra, India

1. Initiated convolutional neural network-based one-shot learning for segregation of plastic waste imagesresulting in 99% accuracy across all plastic waste categories.2. Devised an algorithm that identified unseen plastic waste image categories with a precision of 0.95.

Calgary, Canada Area

1. Set up an image processing-based algorithm for automatic selection and tracking of a region of interest to remove background noise and capture the subtle body movements due to breathing.2. Combined pose estimation and motion detection to identify breathing rate from infra-red cameras, which was integrated in a system designed to monitor the health of homeless people in shelters of Canada.

Mumbai Area, India

1. Worked with a team of interns to design a large-scale ensembled facial recognition system, whichreplaced the identity card-based authorization system, leading to a faster employee entry/exit process2. Implemented detection and removal of outliers to improve the accuracy of the facial recognition system by 15% and simultaneously reduce the memory consumption by 10%.

Master'S Degree, Computer Science

Fall '21 CSE 252A - Computer Vision 1 CSE 258 - Recommender Systems CSE 250A - Probabilistic Reasoning and Learning Winter '22 CSE 251B.

M.S.C Mathematics And B.E Computer Science, Computer Science