Adarsh Modh

I am a Master's Student at the University of Pennsylvania in the Electrical and Systems Engineering Department. My interests broadly lie at the intersection of Robotics, Computer Vision and Machine Learning. Recently, I have been focussed on specific problems in Geometric Computer Vision and Robot Learning.

At Penn, I am a Graduate Researcher at the GRASP lab advised by Dr. Kostas Daniilidis. Currently, I am working on the Smart Aviary Project trying to solve problems in Mult-view geometry, 3D Pose/Shape Estimation, Tracking and Re-identification.

Prior to starting my Master's, I was a Research Assistant at the Robotics Research Center, IIIT-Hyderabad, where I worked on problems like Motion Planning and Control for a Level 3 Autonomous Driving Car. I did my Bachelor's in Electrical Engineering from S. V. National Institute of Technology, India.

Email  /  Resume  /  LinkedIn  /  GitHub  /  Google Scholar

profile photo
fast-texture 3D Bird Reconstruction: a Dataset, Model, and Shape Recovery from a Single View
Marc Badger, Yufu Wang, Adarsh Modh, Ammon Perkes, Nikos Kolotouros, Bernd Pfrommer, Marc Schmidt, Kostas Daniilidis

European Conference on Computer Vision (ECCV), 2020
arXiv  /  project page  /  bibtex

We introduce a model and multi-view optimization approach, which we use to capture the unique shape and pose space displayed by live birds. We then introduce a pipeline and experiments for keypoint, mask, pose, and shape regression that recovers accurate avian postures from single views.

fast-texture Gradient Aware - Shrinking Domain based Control Design for Reactive Planning Frameworks
Adarsh Modh, Siddharth Singh, A. V. S. Sai Bhargav Kumar , Sriram N. N. , K. Madhava Krishna

Proceedings of the Advances in Robotics 2019
arXiv  /  project page  /  bibtex

In this paper, we present a novel control law for longitudinal speed control of autonomous vehicles. The key contributions of the proposed work include the design of a control law that reactively integrates the longitudinal surface gradient of the road into its operation.

fast-texture A Novel Lane Merging Framework with Probabilistic Risk based Lane Selection using Time Scaled Collision Cone
A. V. S. Sai Bhargav Kumar , Adarsh Modh, Mithun Babu ,Bharath Gopalakrishnan , K. Madhava Krishna
IEEE Intelligent Vehicles Symposium (IV) 2018
project page  /  bibtex

In this paper, we present a motion planning framework for autonomous vehicles to perform merge maneuver in dense traffic. Our framework is divided into a two-layer structure, Lane Selection layer and Scale Optimization layer. The Lane Selection layer computes the likelihood of collision along the lanes. This likelihood represents the collision risk associated with each lane and is used for lane selection. Subsequently, the Scale Optimization layer solves the time scaled collision cone (TSCC) constraint re- actively for collision-free velocities. Our framework guarantees a collision-free merging even in dense traffic with minimum disruption.


Updated on Aug 2020, Credits.