Bio

I am a sixth year Ph.D. student in the Electrical and Computer Engineering (ECE) department at Purdue University, advised by professor Saurabh Bagchi. My research interests lie in the joint discipline of Computer Vision and Systems. Particularly, I am working on an adaptive object detection systems for mobile and embedded systems. I received the B.E. degree from Tsinghua University in 2016, where I was with the department of Electronic Engineering and majored in Electronic Information Science and Technology. I did undergraduate thesis with professor Yong Li and researched on topology graphs in Vehicular Networks.

News

I am at the final stage of my Ph.D. program and will graduate soon. I have decided my first job after graduation and thus do not seek jobs in the market.

Recently, I have two papers accepted by TOSN and EMDL and a US patent filed. The ApproxNet paper, which was accepted by TOSN, will be presented on SenSys 2021.

Publications

[C] stands for conference publications. [J] stands for journal publications. [P] stands for patents.

[C1] VideoChef: Efficient Approximation for Streaming Video Processing Pipelines
[Paper] [Presentation Slides] [Presentation Audio] [Poster] [bibtex]
Ran Xu, Jinkyu Koo, Rakesh Kumar, Peter Bai, Subrata Mitra, Sasa Misailovic, and Saurabh Bagchi
In the 2018 USENIX Annual Technical Conference (USENIX ATC 18).
(76 out of 378 submissions, acceptance ratio: 20.1%)

[C2] Pythia: Improving Datacenter Utilization via Precise Contention Prediction for Multiple Co-located Workloads
[Paper] [Presentation Slides] [Presentation Video] [Code] [bibtex]
Ran Xu, Subrata Mitra, Jason Rahman, Peter Bai, Bowen Zhou, Greg Bronevetsky, and Saurabh Bagchi
In Proceedings of the 19th International Middleware Conference (Middleware 18).
(22 out of 95 submissions, acceptance ratio: 23.2%)

[C3] JANUS: Benchmarking Commercial and Open-Source Cloud and Edge Platforms for Object and Anomaly Detection Workloads
[Paper] [Presentation Video] [bibtex]
Karthick Shankar, Pengcheng Wang, Ran Xu, Ashraf Mahgoub, and Somali Chaterji
The IEEE International Conference on Cloud Computing (CLOUD 20)
(35 out of 205 submissions, acceptance ratio: 17.1%)

[C4] ApproxDet: Content and Contention-Aware Approximate Object Detection for Mobiles
[Paper] [Short Video] [Demo Video] [Presentation Video] [Project] [Code] [bibtex]
Ran Xu, Chen-lin Zhang, Pengcheng Wang, Jayoung Lee, Subrata Mitra, Somali Chaterji, Yin Li, and Saurabh Bagchi
The 18th ACM Conference on Embedded Networked Sensor Systems (SenSys 20)
(44 out of 213 submissions, acceptance ratio: 20.7%)

[C5] Closing-the-Loop: A Data-Driven Framework for Effective Video Summarization
[Paper] [Presentation Video] [Project] [bibtex]
Ran Xu, Haoliang Wang, Stefano Petrangeli, Viswanathan Swaminathan, and Saurabh Bagchi
The 22nd IEEE International Symposium on Multimedia (ISM 20)
(16 out of 55 submissions, acceptance ratio: 22.2%)

[C6] Benchmarking Video Object Detection Systems on Embedded Devices under Resource Contention
[Paper] [Presentation Video] [bibtex]
Jayoung Lee, Pengcheng Wang, Ran Xu, Noah Weston, Venkat Dasari, Yin Li, Saurabh Bagchi, and Somali Chaterji
The 5th International Workshop on Embedded and Mobile Deep Learning (EMDL 21)
(7 out of 10 submissions, acceptance ratio: 70.0%)

[J1] On the Opportunistic Topology of Taxi Networks in Urban Mobility Environment
[Paper] [bibtex]
Ran Xu, Yong Li, and Sheng Chen
In IEEE Transactions on Big Data (Volume: 6, Issue: 1, March 1 2020).

[J2] New Frontiers in IoT: Networking, Systems, Reliability, and Security Challenges
[Paper] [bibtex]
Saurabh Bagchi, Tarek F. Abdelzaher, Ramesh Govindan, Prashant Shenoy, Akanksha Atrey, Pradipta Ghosh, and Ran Xu
IEEE Internet of Things Journal (Volume: 7, Issue: 12, Dec. 2020).

[J3] ApproxNet: Content and Contention-Aware Video Object Classification System for Embedded Clients
[Paper] [Short Video] [Presentation Video] [Project] [Demo] [bibtex]
Ran Xu, Rakesh Kumar, Pengcheng Wang, Peter Bai, Ganga Meghanath, Somali Chaterji, Subrata Mitra, and Saurabh Bagchi
ACM Transactions on Sensor Networks (Volume: 18, Issue: 1, Feb. 2022).
Oral presentation on SenSys 2021 (8 out of 49 submissions, acceptance ratio: 16.3%).

[P1] Enhancing media content effectiveness using feedback between evaluation and content editing
[Patent] [bibtex]
Haoliang Wang, Viswanathan Swaminathan, Stefano Petrangeli, and Ran Xu
US Patent 11,170,389, 2021/11/9.

CV

[CV.pdf] (Last updated in 2/2021)

Contact

Email: xu943@purdue.edu
Skype: martin.xuran
Office: Room 34, EE Building, Purdue University - West Lafayette