publications

For a list of my featured publications, please scroll down the home page. This publication page lists all of my major publications. For an exhaustive list that includes workshop papers, please see my CV.

* indicates that authors contributed equally

2022



  1. A didactic figure explaining our core algorithm, and showing some environments we use for evaluation.
    Rethinking Optimization with Differentiable Simulation from a Global Perspective
    Rika Antonova*, Jingyun Yang*,  Krishna Murthy Jatavallabhula, and Jeannette Bohg
    arXiv 2022
  2. Sample object detection (and uncertainty estimates) from our approach.
    f -Cal: Calibrated aleatoric uncertainty estimation from neural networks for robot perception
    Dhaivat Bhatt*, Kaustubh Mani*, Dishank Bansal, Hanju Lee,  Krishna Murthy Jatavallabhula, and Liam Paull
    ICRA 2022
  3. A few representative results from our system - DRACO.
    DRACO: Weakly supervised dense reconstruction and canonicalization of objects
    Rahul Sajnani*, Aadil Mehdi Sanchawala*,  Krishna Murthy JatavallabhulaSrinath Sridhar, and Madhava Krishna K
    ICRA 2022

2021



  1. A sample 3D scene graph of a small house containing multiple rooms.
    Taskography: Evaluating robot task planning over large 3D scene graphs
    Chris Agia*,  Krishna Murthy Jatavallabhula*, Mohamed Khodeir, Ondrej Miksik, Vibhav VineetMustafa MukadamLiam Paull, and Florian Shkurti
    CoRL 2021
  2. A few physical systems implemented in gradSim.
    gradSim: Differentiable simulation for system identification and visuomotor control
    ICLR 2021

2020



  1. An overview of the various components of gradSLAM
    gradSLAM: Dense SLAM meets automatic differentiation
    Krishna Murthy JatavallabhulaGanesh Iyer, and Liam Paull
    ICRA 2020
  2. A preview of the AutoLay benchmark for estimating the bird’s-eye view layout of a road scene.
    AutoLay: Benchmarking Monocular Layout Estimation
    Kaustubh Mani*, Sai Shankar*,  Krishna Murthy Jatavallabhula, and Madhava Krishna K
    IROS 2020
  3. Bird’s-eye view layouts estimated by MonoLayout.
    MonoLayout: Amodal scene layout from a single image
    Kaustubh Mani, Swapnil Daga, Shubhika Garg, Sai ShankarKrishna Murthy Jatavallabhula, and Madhava Krishna K
    WACV 2020
  4. An illustration of our pipeline involving shape and pose optimization of road vehicles.
    Multi-object monocular SLAM for dynamic environments
    Gokul Nair, Swapnil Daga, Rahul Sajnani, Anirudha Ramesh, Junaid Ahmed AnsariKrishna Murthy Jatavallabhula, and Madhava Krishna K
    Intelligent Vehicles Symposium (IV) 2020

2019



  1. An image of the autonomous car (Toyota Prius V) used for MapLite.
    MapLite: Autonomous intersection navigation without detailed prior maps
    Teddy Ort,  Krishna Murthy Jatavallabhula, Rohan Banerjee, Sai Krishna GottipatiDhaivat Bhatt, Igor Gilitschenski, Liam Paull, and Daniela Rus
    IEEE RAL 2019
  2. A splash figure showing various features supported by the Kaolin library.
    Kaolin: A PyTorch Library for Accelerating 3D Deep Learning Research
    Krishna Murthy Jatavallabhula*, Edward Smith*, Jean-Francois Lafleche, Clement Fuji Tsang, Artem Rozantsev, Wenzheng Chen, Tommy Xiang, Rev Lebaredian, and Sanja Fidler
    Whitepaper 2019
  3. An image of the localization experiments (in simulation and on a real robot).
    Deep Active Localization
    Sai Krishna Gottipati*, Keehong Seo*,  Krishna Murthy JatavallabhulaDhaivat Bhatt, Vincent Mai, and Liam Paull
    IEEE RAL 2019
  4. Trajectories forecast in bird’s-eye view by INFER.
    INFER: INtermediate representations for FuturE pRediction
    Shashank Srikanth, Junaid Ahmed AnsariKarnik Ram, Sarthak Sharma,  Krishna Murthy Jatavallabhula, and Madhava Krishna K
    IROS 2019

2018



  1. An image showing vehicles on steep and graded roads being reconstructed as wireframes, on synthetic as well as real datasets.
    The Earth ain’t Flat: Monocular Reconstruction of Vehicles on Steep and Graded Roads from a Moving Camera
    Junaid Ahmed Ansari*, Sarthak Sharma*, Anshuman Majumdar,  Krishna Murthy Jatavallabhula, and Madhava Krishna K
    IROS 2018
  2. CalibNet architecture (block diagram) for geometrically-supervised lidar-camera extrinsic calibration.
    CalibNet: Self-Supervised Extrinsic Calibration using 3D Spatial Transformer Networks
    Ganesh IyerKarnik RamKrishna Murthy Jatavallabhula, and Madhava Krishna K
    IROS 2018
  3. Neural network architecture for geometrically-supervised visual odometry prediction.
    Geometric Consistency for Self-Supervised End-to-End Visual Odometry
    Ganesh Iyer*,  Krishna Murthy Jatavallabhula*, Gunshi GuptaMadhava Krishna K, and Liam Paull
    CVPR Workshops 2018
  4. Sample multi-object tracking results from our approach when evaluated on the KITTI dataset.
    Beyond Pixels: Leveraging Geometry and Shape Cues for Multi-Object Tracking
    Sarthak Sharma*, Junaid Ahmed Ansari*,  Krishna Murthy Jatavallabhula, and Madhava Krishna K
    ICRA 2018
  5. Representative object-SLAM results from the system presented in this paper.
    Constructing Category-Specific Models for Monocular Object SLAM
    Parv Parkhiya, Rishabh Khawad,  Krishna Murthy JatavallabhulaMadhava Krishna K, and Brojeshwar Bhowmick
    ICRA 2018

2017



  1. Car wireframes estimated by the method presented in this paper.
    Shape Priors for Real-Time Monocular Object Localization in Dynamic Environments
    Krishna Murthy Jatavallabhula, Sarthak Sharma, and Madhava Krishna K
    IROS 2017
  2. Car wireframes estimated by the method presented in this paper.
    Reconstructing Vehicles From a Single Image: Shape Priors for Road Scene Understanding
    Krishna Murthy JatavallabhulaSai Krishna Gottipati, Falak Chhaya, and Madhava Krishna K
    ICRA 2017

2016



  1. JIRS
    FAST: Frontier Allocation Synchronized by Token-passing
    Avinash Gautam, Bhargav Jha, Gourav Kumar,  Krishna Murthy Jatavallabhula, Arjun Ram S P, and Sudeept Mohan
    Springer Journal of Intelligent Robotic Systems 2016

2015



  1. SMC
    Cluster, Allocate, Cover: An Efficient Approach for Multi-robot Coverage
    Avinash Gautam,  Krishna Murthy Jatavallabhula, Gourav Kumar, Arjun Ram S P, Bhargav Jha, and Sudeept Mohan
    IEEE SMC 2015
  2. UKSIM
    Maxxyt: An Autonomous Wearable Device for Real-Time Tracking of a Wide Range of Exercises
    Danish Pruthi, Ayush Jain,  Krishna Murthy Jatavallabhula, and Puneet Teja
    UKSIM 2015