I obtained my BA and MEng in Information and Computer Engineering from the University of Cambridge, graduating with a Distinction. Whilst studying at Cambridge, I completed modules in Statistical Pattern Processing, Digital Filters and Spectrum Estimation, Computer Vision and Robotics and Image Processing and Coding. Notably, my Masters project (2013 - 2014) involved writing a program to automate electron gun alignment for a Scanning Electron Microscope (SEM), using GPGPU programming and image processing techniques. I joined the Department of Electronic and Electrical Engineering at University College London (UCL) in 2014, where I'm currently working towards a PhD in the area of image and video indexing and retrieval under the supervision of Dr. Yiannis Andreopoulos and in conjunction with the British Academy of Film and Television Arts (BAFTA) Media Technology.
My research goal is to develop robust and scalable methods for video indexing, similarity detection and classification using deep learning methods and compaction techniques. The aim is to achieve accuracy and efficiency that rivals the best systems available today in these areas.
- A. Chadha and Y. Andreopoulos, "Voronoi-based Compact Image Descriptors: Efficient Region-of-Interest Retrieval With VLAD And Deep-Learning-based Descriptors", IEEE Trans. Multimedia, accepted with minor revision, 2016
- A. Chadha and Y. Andreopoulos, "Region-of-Interest Retrieval in Large Image Datasets with Voronoi VLAD", ICVS Int. Conf. Comput. Vis. Sys., 2015
- Email Confidentiality Classification System: Software deliverable as consultancy work for iProov Ltd, Vasocise Project, funded by Innovate UK. Designed and coded a CNN-based system in TensorFlow that classifies emails as being confidential or non-confidential based on their content.
- Video Indexing and Retrieval System: Software deliverable for BAFTA Media Technology, developed as part of PhD Studentship and Innovate UK VideoClarity Project. Designed and coded two-component retrieval system in MATLAB that extracts content based signatures from a large corpus of images/videos and returns matches to image/video queries from the corpus.
- Royal Commission for the Exhibition of 1851: Industrial Fellowship, ~8 such scholarships awarded annually in the UK
- David Thompson Scholarship: Award for achieving Distinction in MEng