Pranav Kadam

Tencent
Senir Researcher
Palo Alto, CA
May 2023 - present

Immersive Media Research Intern
Palo Alto, CA
February 2023 - May 2023

Supervisors - Xiaozhong Xu, Shan Liu
Project - Mesh Compression

Sony
Applied Research Intern
San Jose, CA
August 2022 - December 2022

Supervisors - Danillo Graziosi, Alexandre Zaghetto, Ali Tabatabai
Project - Dynamic dense point cloud compression, rate control in neural network based point cloud compression

InterDigital, Inc.
Research Intern
New York, NY
May 2022 - August 2022

Supervisors - Dong Tian, Jiahao Pang
Project - Dynamic LiDAR point cloud compression using deep learning techniques, intra-/inter-mode decision in dynamic point cloud compression

University of Southern California
Research Assistant
Los Angeles, CA
May 2019 - March 2023

Supervisor - C.-C. Jay Kuo
Projects - Point cloud classification, registration, odometry, scene flow estimation and pose estimation

University of Southern California
Teaching Assistant
Los Angeles, CA
August 2021 - present

EE 559 - Machine Learning I, Spring 2023
EE 569 - Introduction to Digital Image Processing, Spring 2022
EE 510 - Linear Algebra, Fall 2021

Book

3D Point Cloud Analysis: Traditional, Deep Learning, and Explainable Machine Learning Methods
Shan Liu, Min Zhang, Pranav Kadam, C.-C. Jay Kuo
Springer Nature, 2021
Buy - Springer | Amazon

An introductory book on 3D Point Clouds aimed for graduate students with basic knowledge of Machine Learning and who wish to get started with point clouds.

Publications

First author publications are highlighted.

	A Tiny Machine Learning Model for Point Cloud Object Classification Min Zhang, Jintang Xue, Pranav Kadam, Hardik Prajapati, Shan Liu, C.-C. Jay Kuo APSIPA Transactions on Signal and Information Processing project page / arXiv Single resolution PointHop with multiple spatial aggregations for 3D point cloud classification.
	PointFlowHop: Green and Interpretable Scene Flow Estimation from Consecutive Point Clouds Pranav Kadam, Jiahao Gu, Shan Liu, C.-C. Jay Kuo APSIPA Transactions on Signal and Information Processing project page / arXiv Decomposes 3D scene flow estimation into vehicle ego-motion compensation and object-wise motion estimation steps.
	S3I-PointHop: SO (3)-Invariant PointHop for 3D Point Cloud Classification Pranav Kadam, Hardik Prajapati, Min Zhang, Jintang Xue, Shan Liu, C.-C. Jay Kuo IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023 project page / arXiv Extends PointHop to allow for classification of arbitrarily rotated 3D objects.
	PCRP: Unsupervised Point Cloud Object Retrieval and Pose Estimation Pranav Kadam, Qingyang Zhou, Shan Liu, C.-C. Jay Kuo IEEE International Conference on Image Processing (ICIP), 2022 project page / arXiv First, a similar pre-aligned point cloud is retrieved from the gallery set. Then, the input object is registered to the retrieved object to estimate its pose.
	GreenPCO: An Unsupervised Lightweight Point Cloud Odometry Method Pranav Kadam, Min Zhang, Jiahao Gu, Shan Liu, C.-C. Jay Kuo IEEE International Workshop on Multimedia Signal Processing (MMSP), 2022 project page \| arXiv Extension of R-PointHop for estimating the trajectory of an object from LiDAR scans.
	R-PointHop: A Green, Accurate and Unsupervised Point Cloud Registration Method Pranav Kadam, Min Zhang, Shan Liu, C.-C. Jay Kuo IEEE Transactions on Image Processing, 2022 project page \| arXiv Point cloud registration of indoor scenes and CAD models.
	GSIP: Green Semantic Segmentation of Large-Scale Indoor Point Clouds Min Zhang, Pranav Kadam, Shan Liu, C.-C. Jay Kuo Submitted to Pattern Recognition Letters, 2021 project page \| arXiv Point cloud indoor scene semantic segmentation method using novel point cloud partitioning and processing.
	Unsupervised feedforward feature (UFF) learning for point cloud classification and segmentation Min Zhang, Pranav Kadam, Shan Liu, C.-C. Jay Kuo IEEE Internation Comference on Visual Communication and Image Processing, 2020 project page \| arXiv Point cloud shape part segmentation method.
	Unsupervised Point Cloud Registration via Salient Points Analysis (SPA) Pranav Kadam, Min Zhang, Shan Liu, C.-C. Jay Kuo IEEE Internation Comference on Visual Communication and Image Processing, 2020 project page \| arXiv Extension of PointHop++ for point cloud registration using salient points derived from local geometry information.
	PointHop++: A lightweight learning model on point sets for 3D classification Min Zhang, Yifan Wang, Pranav Kadam, Shan Liu, C.-C. Jay Kuo IEEE International Conference on Image Processing, 2020 project page \| arXiv Incorporation of Channel-wise Saab transform for efficient point cloud classification
	PointHop: An Explainable Machine Learning Method for Point Cloud Classification Min Zhang, Haoxuan You, Pranav Kadam, Shan Liu, C.-C. Jay Kuo IEEE Transactions on Multimedia, 2020 project page \| arXiv Point cloud classification method based on Successive Subspace Learning (SSL)

Miscellaneous

Reviewer, IEEE International Conference on Image Processing (ICIP), 2022
Reviewer, Springer Nature
Reviewer, APSIPA Transactions on Signal and Information Processing
Reviewer, ISPRS Journal of Photogrammetry and Remote Sensing
Course Grader, EE 569 - Introduction to Digital Image Processing, Spring 2020
Course Grader, EE 562 - Random Processes, Spring 2020
Course Grader, EE 567 - Communication Systems, Fall 2019
Student Tech Assistant, USC Transportation, Jan 2019 - Jun 2019

Collaborated with Dipankar Bhattarcharya from IUCAA, Pune on research on detection and analysis of Gamma Ray Bursts (GRBs) using Machine Learning techniques
Collaborated with Dr. Arti Khaparde and Indian Space Research Organization (ISRO) in project on study of vegetation dynamics from satellite images
Interned for part-time at Anveshak Technology and Knowledge Solutions, Pune in IoT domain
Led in organizing conferences, workshops, guest lectures, and competitions as a part of MIT Pune IEEE Students Branch
Arranged study tours and other recreation activities for students

Source code by Jon Barron.