People and Topics

Note

If you are interested in joining CAM2 during Summer 2020, please refer to the New Member Onboarding page to apply.

cam2logo

These projects form the CAM2 (Continuous Analysis of Many CAMeras) family.

2020 Spring Schedule

All meetings are held in EE 013.

Time

Monday

Tuesday

Wednesday

09:00-10:00

10:00-11:00

Software Engineering

11:00-12:00

12:30-13:30

Forest

13:30-14:30

Drone

Human Behavior

14:30-15:30

Embedded 1

Crowdsourcing

15:30-16:30

Leaders

16:30-17:30

Embedded 2

17:30-18:20

VIP lecture (EE 129)

Due to the large number of team mebers, it is not possible changing the regular meeting time. If your schedule does not fit into a particular team, you have to move to another team.

Topics

Low-Power Computer Vision

Many applications of computer vision runs on battery-powered systems, susch as mobile phones, drones, IoT (Internet of Things), and edge devices. This project investigates how to reduce the energy consumption of computer vision. The project uses many methods to achieve better energy efficiency, such as

  • Restructure neural networks. Commonly adopted neural networks are large complex models. Many connections among neurons, however, are not necessary and can be removed. By restructuring the networks, it is possible avoiding large amounts of compuation and save energy.

  • Reduce the input space. Many machine learning models are designed for general purposes and can recognize objects of hundreds of classes. Many applications, however, have only a few classes of objects. For example, it is not possible seeing an elephant in a classroom. Thus, the machine models do not have to detect elephants.

  • Utilize hardware features. Transform the mathematical equations for machine models in order to map to hardware features (such as memory hierarchies) efficiently.

This project is supported by Facebook, Google, and Xilinx.

Readings for new members:

Analysis of Drone Video

This project creates computer vision solutions recognizing objects captured by cameras mounted on drones. In Fall 2019, the team will create a set of video clips for the following purposes:

  • Construct three-dimensional geometries of objects: The video clips will capture cardboard boxes of different sizes, together with a wide range of objects and several with known sizes.

  • Detect and track multiple moving objects: The clips include moving objects. The drone itself is also moving. The purpose is to correctly identify these objects and track their movements.

  • Segmentation: Create pixel-wise labels of different objects.

  • Re-identify people: Determine whether the same person has been before.

This project is supported by NSF CNS-1925713

Readings for new members:


Analyze Human Behavior in Video

The purpose of this team is to use real-time video analytics to detect dangerous behavior or safety violation in workplace (such as factories), raise alerts to prevent injury, or provide post-event analysis to prevent future occurrences. In Fall 2019, the team will focus on solving these problems in an indoor environment with multiple cameras:

  • Where are the people (including re-identifying the same person in different cameras)?

  • Where does each person face?

Readings


Embedded Vision 1

Recent progress in computer vision has focused primarily in general-purpose object detection using datasets with many (hundreds) categories of objects (such as humans, dogs, vehicles, furniture, buildings, etc.). For many applications, however, the number of possible objects can be limited. For example, inside an airport terminal, elephants or eagles are not expected. This project will use computer graphics to synthesize images and videos of these scenarios. The synthesized data is used to train computer vision running on embedded systems (also called edge devices). Doing so can reduce network traffic and make the system more scalable. Moreover, sensitive information (such as human faces) may be detected and protected before the data leaves the cameras.

Readings for new members:

Embedded Vision 2

This project investigates computer vision solutions that can perform the following tasks in an embedded computer (small enough to be inside a typical camera)

  • Obtain aggregate information (such as the number of people and their genders)

  • Detect faces

  • Encrypt the faces before sending the data to storage

The sensitive data (faces) never leaves the camera. Only authorized people with the decryption key can see the faces. The concept is illustrated below.

embeddedprivacy

Readings for new members:


Forest Inventory Analysis

Forest inventory analysis is time consuming and expensive to do manually. The team is researching the use of stereo cameras and video footage to obtain individual tree information efficiently and at a low cost. Our goals this semester are as follows: - Generate a 3D reconstruction of a forest plot from video footage - Uniquely identify each tree in the 3D reconstruction with a diameter. - Compute the taper of a tree using stereo video - Compute the crown height of a tree using stereo video

The following images show the result from a distance sensor and the tree image (before and after denoising).

forest03 forest04 forest05

The following images are snapshots of 3D digital reconstructions of trees from video.

forest06

Readings for new members:


Software Engineering for Machine Learning

This project creates a process for developing reproducible software used in machine learning. In Fall 2019, the team’s focus is to create tools that faciliate code review. The tools analyze the histories of version control repositories and automatically identify possible defects within a pull request. The tools will also collect metrics relating to the code review.

Readings


Crowdsourcing

Crowdsourcing utilizes the knowledge of humans to complete a task. In the scope of research, we are using the crowd to handle specific tasks that may be hard for a machine to do or improve the work of a machine. This may include tasks such as detecting bias within image datasets using human knowledge rather than machines, since humans are better at distinguishing features within images. This semester, we are utilizing crowdsourcing to help select the most suitable machine model to use for unsupervised domain adaptation. We use the crowd to classify images to datasets and generate a confusion matrix detailing the similarity of images across several datasets.

Reading for new members:

crowdsource03 crowdsource02

crowdsource05 crowdsource04

Faculty

https://ag.purdue.edu/ProfileImages/dbarbara.jpg

David Michael Barbarash

Landscape Architecture, Purdue

https://engineering.purdue.edu/ResourceDB/ResourceFiles/image92690

Dave Cappelleri

Mechanical Engineering, Purdue

https://shuohanchen.files.wordpress.com/2019/02/shuohan-eps-converted-to.png?w=220&h=300

Shuo-Han Chen

Institute of Information Science, Academia Sinica

https://drive.google.com/uc?id=1EqxgXBuEQNiQ5pNVvg42AfWMFKByjKh1

Yung-Hsiang Lu

Electrical and Computer Engineering, Purdue

http://www.stat.purdue.edu/images/Faculty/thumbnail/varao-t.jpg

Vinayak Rao

Statistics, Purdue

https://drive.google.com/uc?id=19_-2sKwLTcjoBvjclB8tqlIA56k5QwUq

Guofan Shao

Professor, Forestry and Natural Resources, Purdue

https://avatars1.githubusercontent.com/u/651504?s=460&v=4

George K. Thiruvathukal

Computer Science, Loyola University Chicago.

http://www.stat.purdue.edu/~mdw/images/WardMFO.jpg

Mark Daniel Ward

Statistics, Purdue

https://ag.purdue.edu/ProfileImages/woeste.jpg

Keith E. Woeste

Forestry and Natural Resources, Purdue

https://www.cs.purdue.edu/people/images/small/faculty/mingyin.jpg

Ming Yin

Computer Science, Purdue

Members

Graduate Students

https://drive.google.com/uc?id=1YunKydNN7OS_vvBubbME4UykfjNh1CgA

Abhinav Goel: Doctoral Student, Improve Neural Networks’ Energy Efficiency

https://drive.google.com/uc?id=1GzpDueX6W2e4sx0OGfKm51cru34jyEvp

Sara Aghajanzadeh: Master Student, Detect Faces and Protect Privacy

https://drive.google.com/uc?id=1kIYIrkXnICIb2odq5WWGlsdCYv4fTpVU

Ryan Dailey: Master Student, Discover Network Cameras

https://drive.google.com/uc?id=1rtNrB-sPUJ6gllceGkgXex4gH11xZBmu

Caleb Tung: Doctoral Student, Using Contextual Information from Network Cameras to Improve and Evaluate Computer Vision Solutions

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/shreya_ghosh.jpg

Shreya Ghosh: Master Student, Data-Driven Civil Infrastructure Decision Making for Improved Citizen Safety, Accessibility, and Economic Opportunity

Undergraduate Students and Spring 2020 Teams

Drone Video

Create datasets of drone video, recognize objects, estimate the sizes.

https://i.ibb.co/4SB910F/20190617-180018.jpg

Avanish Subbiah

Leader

https://drive.google.com/uc?id=1BgdG9XYcrmdMtdSbePpp324jwdnwl_7p

Xiao Hu

Co-leader

https://drive.google.com/uc?id=1Yv1UB8v2LEjDsRCaCDilNzI68hoDAOXf

Katherine Sandys

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_justin_qualley.png

Justin Qualley

https://drive.google.com/uc?id=1M3PTWnGPrRkv2ntr_Furr-gh1VezKSL_

Victor Oduduabasi

https://drive.google.com/uc?id=1t-krvZinKrSk1YT8MRl8R6xoPUHpF8H7

Haobo Wang

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_placeholder_male.jpg

Nathan Gizaw

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_placeholder_male.jpg

Karthik Maiya

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_placeholder_male.jpg

Rushabh Ramesh Ranka

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_tuhin_sarkar.jpg

Tuhin Sarkar

https://drive.google.com/uc?id=1YunKydNN7OS_vvBubbME4UykfjNh1CgA

Abhinav Goel

Embedded Vision 1

Detect and encrypt faces to protect privacy using embedded computer.

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_vaastav_arora.jpg

Vaastav Arora

Leader

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_siddharth_srinivasan.jpg

Siddharth Srinivasan

Co-leader

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_rufat_imanov.jpeg

Rufat Imanov

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_placeholder_female.jpg

Asa Cutler

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_xin_wang.jpeg

Xin Wang

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_xin_du.jpeg

Xin Du

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_placeholder_male.jpg

Seyram Samuel Mortoti

Embedded Vision 2

Identify the specific features (called distinctiveness) of different visual dataset. Use one dataset with many labels to help train machine models for another datasets with few labels.

https://drive.google.com/uc?id=1db-0VtllDb1sU5MeqnmOT9WKNMYsPy5z

Li Yon Tan

Leader

https://drive.google.com/uc?id=1RbbxJV5LEmUpPmfMcUU9V457Ww4wtN2t

Isha Ghodgaonkar

Co-leader

https://drive.google.com/uc?id=1TfZn-I0Mk_lvY-cMontY8f1u9o5Zvc-y

Akshay Pawar

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_placeholder_female.jpg

Damini Rijhwani

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_placeholder_male.jpg

Hojoung Jang

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_aditya_chakraborty.jpeg

Aditya Chakraborty

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_shristi_saraff.jpg

Shristi Saraff

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_placeholder_female.jpg

Meenakshi Pavithran

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_zach_berg.jpg

Zach Berg

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_jackson_moffet.jpg

Jackson Moffet

https://drive.google.com/uc?id=1rtNrB-sPUJ6gllceGkgXex4gH11xZBmu

Caleb Tung

Doctoral Student

Crowdsourcing for Data Bias

Use crowdsourcing to identify unintended bias in visual data and label data.

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_gore_kao.jpg

Gore Kao

Leader

https://drive.google.com/uc?id=1Pik3biv3epBSKMobHHiqH8FcG9679ZSu

Kaiwen Yu Co-leader

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_yukyung_lee.jpg

Yukyung Lee

https://drive.google.com/uc?id=1t-krvZinKrSk1YT8MRl8R6xoPUHpF8H7

Haobo Wang

https://drive.google.com/uc?id=1BgdG9XYcrmdMtdSbePpp324jwdnwl_7p

Xiao Hu

CAM² Co-leader

https://drive.google.com/uc?id=1u5dbejyw-62y5x6UPKEtPo3DFd4AtYCc

Anirudh Vegesana

https://drive.google.com/uc?id=1yUr73JBTlTG0LMew8pqVXA5csNggmuOX

Ashley Kim

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_esteban_gorostiaga.png

Esteban Gorostiaga

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_phillip_archuleta.jpeg

Phillip Andrew Archuleta

https://drive.google.com/uc?id=1Y7Ew3_OUSKDagQ7nOkRKOpYV1DhrPdi7

Fischer Bordwell

https://www.cs.purdue.edu/people/images/small/faculty/mingyin.jpg

Ming Yin

Advisor

Forest Inventory

Use computer vision to recognize tree species and estimate their sizes.

https://i.ibb.co/y5GW8Hk/Selfie-2.jpg

David Jarufe

Leader

https://drive.google.com/uc?id=1WrLZtXkzgHDQbCC0XLX92C8a8rgS6yMd

Yezhi Shen

CAM² Leader

https://drive.google.com/uc?id=1GeeVgSnl4Fwf-rlIFlG5LuSohcMMIpTi

Nick Eliopoulos

CAM² Co-leader

https://i.ibb.co/Jxfdyb8/IMG-9250.jpg

Ya Ling Tsai

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_justin_hsiung.png

Yi-Fang Hsiung

Human Behavior in Video

Understand human behavior and obtain aggregate information from video.

https://drive.google.com/uc?id=15MjIOEr5k1XvctTmnAM3yzy3JduC0MJm

Taher Dohadwala

Leader

https://avatars3.githubusercontent.com/u/20303922?s=460&v=4

Moiz Rasheed

Co-leader

https://drive.google.com/uc?id=1wzCPewNOxMyl3QBXAuwVUN_jeLqL8hnK

Wenxi Zhang

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_ethan_glaser.png

Ethan Glaser

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_nour_hendy.jpeg

Noureldin Hendy

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_siddhartha_kumar.jpg

Siddhartha Kumar Senthil Kumar

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_mert_zamir.jpg

Mert Zamir

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_placeholder_male.jpg

Tong Wang

https://ag.purdue.edu/ProfileImages/dbarbara.jpg

David Michael Barbarash

Advisor

https://drive.google.com/uc?id=1GzpDueX6W2e4sx0OGfKm51cru34jyEvp

Sara Aghajanzadeh

Masters Student

Software Engineering

Create procedure for developing high-quality and reproducible software.

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_akhil_channakotla.jpg

Akhil Chinnakotla

Leader

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_connor_chadwick.jpg

Connor Chadwick

https://i.postimg.cc/VkM4Gjjj/ezgif-com-crop.jpg

Ryan Firestone

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_stephen_davis.jpg

Stephen Davis

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_seoyoung_lee.jpg

Seoyoung Lee

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_amogh_shanbag.jpg

Amogh Shanbag

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_ryan_chen.jpg

Ryan Chen

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_placeholder_male.jpg

Ved Dave

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_jack_lecroy.jpg

Jack LeCroy

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_rohit_tokala.jpg

Rohit Reddy Tokala

https://drive.google.com/uc?id=1p5Ad8doE42SN57LXYTdsDtKOjmgGbPfG

Noah Curran

https://raw.githubusercontent.com/PurdueCAM2Project/HELPSweb/master/source/images/member_vishnu_banna.jpeg

Vishnu Banna

https://avatars1.githubusercontent.com/u/651504?s=460&v=4

George K. Thiruvathukal

Advisor

Alumni

Achinthya Soordelu

James Lee

He Li

Anthony Fennell

Jenil Patel

Ehren Marschall

Fengjian Pan

Nirmal Asokan

Sanghyun Cho

Shengli Sui

Woojin Kim

Ajay Gopakumar

Jiancheng Wang

Sitian Lu

Juncheng Tang

Milos Malesevic

Mina Guo

Hanyang Liu

Zhenming Zhang

Zaiwei Zhang

Jiaju Yue

Huanyi Guo

Jeanne Deng

Zhenming Zhao

Anthony Kang

Qingshuang Chen

Yuhao Chen

Borui Chen

Sriram Rangaramanujan

James Tay

Kyle Mcnulty

Seth Bontrager

Pranjit Kalita

Subhav Ramachandran

Everett Berry

Erik Rozolis

Bolun Zhang

Andrew Green

Yukun An

Daniel Dilger

Yexin Wang

Zhifan Zeng

Joseph Sweeney

Ryan Schlueter

John Laiman

Jay Patel

Yutong Huang

Yuxiang Zi

Zhanxiang Hua

Weizhi Li

Yash Pundlik

Ramyak Singh

Nanxin Jin

Kyle Martin

Hao Zou

Sam Yellin

Wenzhong Duan

Aparna Pidaparthi

Changyu Li

Deepika Aggrawal

Hanwen Huang

Hussni Mohd Zakir

Sihao Yin

Weiqing Huang

Christopher Jovanovic

Ahmed Kaseb

Wengyan Chan

Meera Haridasa

Deeptanshu Malik

Vadim Nikiforov

Matthew Fitzgerald

Youngsol Koh

Mehmet Alp Aysan

Cailey Farrell

Yifan Li

Lucas Neumann

Robert Gitau

Zhi Kai Tan

Spencer Huston

Mohamad Alani

Lucas Wiles

Yuxin Zhang

Chau Minh Nguyen

Shunqiao Huang

Minh Nguyen

Dhruv Swarup

Ryan Dailey

Video by Current and Former Members