Saket Anand
Asst. Professor
B-103, Academic Block
Tel:+91-11-2690 7425

Current Students

PhD Students

M.Tech. Students (Thesis)

  • Poojankumar Oza
  • Aditay Tripathi

Prospective Students

I am presently looking for highly motivated students to work in the following areas during Summer 2017 and beyond:

  • Vision based Perception for Driver Assistance
  • The following problems are of interest to this project

    • Lane Detection
    • Road Surface Segmentation
    • Traffic Light Detection
    • Road Sign Detection
    • Obstacle Detection and Classification
    • Pedestrian Detection

    The objective is to develop robust algorithms for the aforementioned problems, which can run in near-real time. We plan to solve these problems using Deep Learning tools and are looking for strong B.Tech. and M.Tech. students who are familiar with deep learning tools like Caffe/TensorFlow/Torch/Theano. If you have excellent programming skills and are excited about computer vision and machine learning, and you are interested in a longer-term association, please send me your CV and availability details in an email to me with the Subject Line:"[2017-Driverless Car Project]".

  • Personalized Text to Speech Synthesis
  • A key requirement in a natural sounding, personalized, speech synthesis system is its ability to mimic the accent and pronunciation of the user. Several approaches have been explored in the past including Synthesis by concatenation, Hidden Markov Models (HMMs) and Unit-Selection based synthesis. Recently, deep learning based approaches have been explored for natural sounding computer generated speech.

    In this project, we wish to understand the underlying principles governing pronunciation, accents and prosody that are key to natural sounding speech. Our goal is to model these latent parameters using advanced machine learning techniques and integrate these models with state-of-the-art speech synthesis tools.

    I am looking for one PhD student with a strong background in Signal Processing and Probability & Statistics and a willingness to learn and apply Machine Learning algorithms. B.Tech. students who are interested in doing a BTP related to this project may contact me. Strong programming skills (C/C++/Python) and good understanding of Signal Processing and/or Machine Learning techniques is required.

Past Students

  • Sarthak Ahuja (B.Tech. Project, 2016, co-advised by Chetan Arora)
  • Gullal Singh Cheema (M.Tech. Thesis, Summer 2016)
  • Shreya Tripathi (M.Tech. Thesis, Summer 2016)