Ankit Dhall

I am a graduate student at ETH-Zurich studying Robotics and am particularly interested in exploring research areas at the intersection of Computer Vision and Machine Learning. I work as a research intern at nuTonomy working on safety-critical perception for autonomous driving.

Currently, I am the Head of Perception at Academic Motorsport Association of Zurich (AMZ) for the Formula Student Driverless, Season 2019. In 2018, I contributed to the Computer Vision package at AMZ Racing for the Driverless competition. We were declared Champions at Formula Student Italy at Varano de Melegari and Formula Student Germany at Hockenheimring, the biggest student engineering competition in the world.

Here's a video of "gotthard driverless" cruising autonomously at the Hockenheimring.

In 2015, I had the opportunity to work with Prof. James Davis and Rajan Vaish during my sophomore year. I spent a wonderful summer of 2016 with Prof. Wolfram Burgard's research group at the University of Freiburg, Germany funded by DAAD. I completed my Bachelor's at VIT University studying Computer Science and my thesis at Prof. K.M. Krishna's lab at Robotics Research Center, IIIT-Hyderabad.

Email  /  CV  /  Google Scholar  /  LinkedIn  /  Github


I'm interested in computer vision, machine learning and robotics.

LiDAR-Camera Calibration using 3D-3D Point correspondences
Ankit Dhall, Kunal Chelani, Vishnu Radhakrishnan, K.M. Krishna
arXiv pre-print, 2017

abstract | arXiv | ROS package | widely forked GitHub repo | videos [1] [2] [3]

AdapNet: Adaptive Semantic Segmentation in Adverse Environmental Conditions
Abhinav Valada, Johan Vertens, Ankit Dhall, Wolfram Burgard
ICRA, 2017

abstract | project page + web demo | videos [1] [2] [3]

Convoluted Mixture of Deep Experts for Robust Semantic Segmentation
Abhinav Valada*, Ankit Dhall*, Wolfram Burgard
State Estimation and Terrain Perception for All Terrain Mobile Robots Workshop at IROS, 2016

abstract | project page + web demo

On Optimizing Human-Machine Task Assignments
Andreas Veit, Michael Wilber, Rajan Vaish, Serge Belongie, James Davis, ..., Ankit Dhall, et al.
HCOMP, 2015

abstract | arXiv

Template credits: Jon Barron.