Aarush Gupta

Aarush Gupta

Research Engineer



I am a research engineer at Rephrase.ai, where I work on developing tech which generates videos from just text. I recently graduated from IIT Roorkee with a bachelor’s degree in Computer Science and Engineering. I am broadly in Deep Learning, with applications in Computer Vision and Natural Language Processing.

During my undergrad days, I was fortunate to work with Prof. Hanspeter Pfister and Dr. Donglai Wei at Harvard University, Prof. Venkatesh Babu’s group at Indian Institute of Science, and Prof. Marco Pedersoli and Prof. Jose Dolz at École de Technologie Supérieure. I was also the Vice-Chair of ACM IIT Roorkee Student Chapter, and co-led a Deep Learning reading group named the Vision and Language Group

Apart from AI, I am enthusiastic about financial markets, guitar, food, and movies.


  • Computer Vision
  • Deep Learning
  • Natural Language Processing


  • B.Tech in Computer Science and Engineering, 2020

    Indian Institute of Technology Roorkee


  • June 2020: “MitoEM Dataset: Large-scale 3D Mitochondria Instance Segmentation from EM Images” accepted at MICCAI 2020.

  • May 2020: Joined Rephrase.ai as a Research Engineer.

  • April 2020: Graduated from IIT Roorkee.

  • December 2019: Joining Rephrase.ai for the winter. Here’s a cool video about their work.

  • April 2019: Awarded the SN Bose Scholarship! Will be visiting Prof. Hanspeter Pfister at Harvard University over the summers.

  • November 2018: Accepted as a research intern at Video Analytics Lab, IISc! Will be visiting Prof. R. Venkatesh Babu over the winters.

  • August 2018: Our paper “An Attention Model for Group Level Emotion Recognition” has been accepted at ICMI 2018!

  • April 2018: Excited to work with Prof. Marco Pedersoli and Prof. Jose Dolz over the summers!


SN Bose Scholarship Awardee

Recipient of the SN Bose scholarship, which aims to nurture future innovators and thought leaders. Joint initiative by the

  • Science & Engineering Board (SERB)
  • Department of Science and Technology (DST)
  • Govt. of India, the Indo-U.S. Science and Technology Forum (IUSSTF)
  • WINStep Forward

to develop a dynamic and transformative student exchange program between premier institutions in India and the United States.

All India Rank 363

  • Achieved a rank of 363 out of 200,000 candidates.
  • JEE Advanced is an entrance exam to get an admit into IITs, the premier engineering institutes in India, with a selection percentage of only 5%.

KVPY Fellowship Awardee

A highly prestigious National Fellowship awarded by Indian Institute of Science and Government of INDIA to students who show high aptitude in research.