Siddharth Choudhary

I am a Senior Applied Scientist in Amazon AGI. I closely collaborate with AWS AI Labs. Before this, I was an applied scientist in Amazon Halo where I worked on problems at the intersection of Computer Vision and Health.

Prior to Amazon, I was a Principal Computer Vision Researcher at Magic Leap. I finished my Ph.D. from the School of Interactive Computing at Georgia Institute of Technology. I was advised by Professor Henrik I. Christensen and Professor Frank Dellaert. I completed my Bachelors and Masters in Computer Science from IIIT Hyderabad where I was advised by Prof. P J Narayanan.

Email  /  CV  /  Google Scholar  /  Twitter  /  LinkedIn  /  Github

profile photo
Research

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

Multi-modal hallucination control by visual information grounding
Alessandro Favero, Luca Zancato, Matthew Trager, Siddharth Choudhary, Pramuditha Perera, Alessandro Achille, Ashwin Swaminathan, Stefano Soatto,
CVPR, 2024  
arXiv / web

We address hallucination in Generative Vision-Language Models by introducing Multi-Modal Mutual-Information Decoding (M3ID), which amplifies the influence of reference images, reducing hallucinated responses by up to 28% without compromising linguistic fluency.

SplatArmor: Articulated Gaussian splatting for animatable humans from monocular RGB videos
Rohit Jena, Ganesh Iyer, Siddharth Choudhary, Brandon M. Smith, Pratik Chaudhari, James C. Gee,
arXiv, 2023  
arXiv / web

A fully articulated Gaussian splatting model for human avatars. Our model includes both rigid and non-rigid skinning components, and a Neural Color Field for implicit color regularization.

Development and Validation of an Accurate Smartphone Application for Measuring Waist-to-Hip Circumference Ratio
Siddharth Choudhary, Ganesh Iyer, Brandon M. Smith, Jinjin Li, Mark Sippel, Antonio Criminisi, Steven B Heymsfield
Nature Digital Medicine, 2023 (Shipped with Amazon Halo Body)
pdf / web / citation

Propose a CNN based model called MeasureNet for accurately and reliably predicting body measurements and waist-hip ratio. Model is trained using realistic synthetic dataset.

Mesh Strikes Back: Fast and Efficient Human Reconstruction from RGB videos
Rohit Jena, Pratik Chaudhari, James C. Gee, Ganesh Iyer, Siddharth Choudhary, Brandon M. Smith
arXiv, 2023  
arXiv

Optimizing a SMPL+D mesh and an efficient, multi-resolution texture representation for novel view synthesis and pose synthesis.

Multiuser, Scalable 3D Object Detection in the AR Cloud
Siddharth Choudhary, Nitesh Sekhar, Siddharth Mahendran Prateek Singhal
CVPR Workshop on AR/VR, 2020   (Shipped on Magic Leap One)
pdf / project page / bibtex

An approach for multiuser and scalable 3D object detection, based on distributed data association and fusion.

Data-Efficient Decentralized Visual SLAM
Titus Cieslewski, Siddharth Choudhary, Davide Scaramuzza
ICRA, 2018  
pdf / video / code / bibtex

Decentralized visual SLAM combining distributed PGO using Gauss-Seidel and efficient, distributed place recognition using NetVLAD features.

Distributed Object based SLAM
Siddharth Choudhary
PhD Thesis, 2017  
pdf / ppt

Decentralized object based SLAM combining distributed PGO with joint modeling and mapping of object landmarks.

Distributed Mapping with Privacy and Communication Constraints: Lightweight Algorithms and Object-based Models
Siddharth Choudhary, Luca Carlone, Carlos Nieto, John Rogers, Henrik I. Christensen, Frank Dellaert
IJRR, 2017  
arxiv / code / video / bibtex

Propose a distributed implementation of the two-stage pose graph optimization, using Successive Over-Relaxation (SOR) and the Jacobi Over-Relaxation (JOR) as workhorses to split the computation among the robots. Extends it to work with object-based map models.

Multi Robot Object-based SLAM
Siddharth Choudhary, Luca Carlone, Carlos Nieto, John Rogers, Zhen Liu, Henrik I. Christensen, Frank Dellaert
ISER, 2016  
pdf bibtex

Multi robot SLAM approach that uses 3D objects as landmarks for localization and mapping. Leverages local computation at each robot (e.g., object detection and object pose estimation) to reduce the communication burden.

Distributed Trajectory Estimation with Privacy and Communication Constraints: a Two-Stage Distributed Gauss-Seidel Approach
Siddharth Choudhary, Luca Carlone, Carlos Nieto, John Rogers, Henrik I. Christensen, Frank Dellaert
ICRA, 2016  
pdf / ppt / www / video / bibtex

Leverages recent results which show that the maximum likelihood trajectory is well approximated by a sequence of two quadratic subproblems and solves it in a distributed manner, using the distributed Gauss-Seidel (DGS) algorithm.

Active planning based extrinsic calibration of exteroceptive sensors in unknown environments
Varun Murali, Carlos Nieto, Siddharth Choudhary, Henrik I. Christensen
IROS, 2016  
pdf / code

Plans a trajectory which actively reduces the uncertainty of the robot's calibration given a rough initial calibration estimate.

Exactly Sparse Memory Efficient SLAM using the Multi-Block Alternating Direction Method of Multipliers
Siddharth Choudhary, Luca Carlone, Henrik I. Christensen, Frank Dellaert
IROS, 2015  
pdf / code / bibtex

Scalable SLAM using multiblock Alternating Direction Method of Multipliers (ADMM).

Information-based Reduced Landmark SLAM
Siddharth Choudhary, Vadim Indelman, Henrik I. Christensen, Frank Dellaert
ICRA, 2015  
pdf / bibtex

Information theoretic algorithm to efficiently reduce the number of landmarks and poses in a SLAM estimate without compromising the accuracy of the estimated trajectory.

SLAM with Object Discovery, Modeling and Mapping
Siddharth Choudhary, Alexander J. B. Trevor, Henrik I. Christensen, Frank Dellaert
IROS, 2014  
pdf / bibtex

Propose an approach for online object discovery and object modeling, and extend a SLAM system to utilize these discovered and modeled objects as landmarks to help localize the robot in an online manner.

CPU and/or GPU: Revisiting the GPU Vs. CPU Myth
Kishore Kothapalli, Dip Sankar Banerjee, P. J. Narayanan, Surinder Sood, Aman Kumar Bahl, Shashank Sharma, Shrenik Lad, Krishna Kumar Singh, Kiran Matam, Sivaramakrishna Bharadwaj, Rohit Nigam, Parikshit Sakurikar, Aditya Deshpande, Ishan Misra, Siddharth Choudhary, Shubham Gupta
arXiv, 2013  
arXiv

Geometry directed browser for personal photographs
Aditya Deshpande, Siddharth Choudhary, PJ Narayanan, Krishna Kumar Singh, Kaustav Kundu, Aditya Singh, Apurva Kumar
ICVGIP, 2012  
pdf / bibtex

Improving the Efficiency of SfM and its Applications
Siddharth Choudhary
MS Thesis, 2012  
pdf / www

Visibility Probability Structure from SfM Datasets and Applications
Siddharth Choudhary, P. J. Narayanan
ECCV, 2012  
pdf / www / bibtex

Encode the visibility information between and among points and cameras as visibility probabilities for improved localization and triangulation.

Practical time bundle adjustment for 3d reconstruction on the gpu
Siddharth Choudhary, Shubham Gupta, P. J. Narayanan
ECCV CVGPU Workshop, 2010  
pdf / ppt / tutorial

Hybrid implementation of sparse bundle adjustment on the GPU using CUDA, with the CPU working in parallel.

Professional Activities
Reviewer for CVPR, ICCV, ECCV, 3DV, ICRA, IROS, TRO, IJRR
Organizer for CVPR 2019 tutorial on Perception at Magic Leap
Patents
Cross Reality System
Scalable three-dimensional object recognition in a cross reality system
Body dimensions from two-dimensional body images
Misc
Literature Review on Minimal Mapping for Navigation
SLAM Literature Review
Bundle adjustment tutorial

Credits to Jon Barron for website template.