This is Bo Chen, a postdoc at University of Illinois at Urbana-Champaign (UIUC) working with Prof. Klara Nahrstedt. I obtained my Ph.D. in computer science at UIUC in 2022, advised by Prof. Klara Nahrstedt. Before coming to UIUC, I got my B.E. degree from Shanghai Jiao Tong University (SJTU), under the supervision of Prof. Xinbing Wang.
Immersive computing enhances human perception by delivering remote data and analyzing it to create richer, more interactive experiences. Recent advancements in Artificial Intelligence (AI) (i.e., neural representations and large models) introduce both opportunities and challenges to immersive computing. On the one hand: 1.Neural representations (e.g., neural radiance fields and neural codecs) greatly advance data delivery with better compactness and photo-realism. 2. Large models (e.g., large language models) boost data analytics with the ability to correlate spatial-temporal and multimodal content. On the other hand, AIβs training and inference demand unprecedented volumes of data to be collected, transmitted, and processed. Under a real-world setting with limited bandwidth, resource-constrained devices, or hazardous environments, AI-integrated immersive computing systems may fail to meet system requirements (e.g., latency and frame rate). Even if they reduce the AI workload via system optimizations that downsample data or sparsify AI models, they inevitably sacrifice AI performance (e.g., accuracy) by ignoring the complex impact of system optimizations on AI models.
My research advocates AI-system co-design, an approach that resolves the complex impact of system optimizations on AI models with two core principles:
- System-for-AI Optimization: Employing system optimization techniques in software and hardware to manage data flow and computing loads while preserving AI performance.
- AI-for-System Customization: Customizing AI components, including neural network architecture, training processes, and inference pipelines, to mitigate the impact of system optimizations, ensuring satisfactory Quality of Experience (QoE) and accuracy of the system.
Guided by these principles, my research agenda bridges the gap between AI-integrated immersive computing systems and the real world with three key contributions: Bandwidth Efficiency, Computation Scaling, and Hazard Resilience**
Research Interests: Networking, operating systems, immersive/visual/spatial computing, and mobile computing.
I am now actively looking for tenure-track faculty positions (2024-2025). Feel free to ping me if you have any openings.
Here is my CV. Contact: boc2 AT illinois DOT edu
π₯ News
- 2024.11: Β ππ Invited Talk at UM-SJTU Joint Institute, Advancing Immersive Computing with AI-System Co-design.
- 2024.09: Β ππ Paper Conditionally Accepted in ACM SenSysβ24, ImmerScope: Multi-view Video Aggregation at Edge towards Immersive Content Services.
- 2024.04: Β ππ Invited Talk at UIUC Sys-Net Spring 2024 Retreat, NeRFHub: A Context-Aware NeRF Serving Framework for Mobile Immersive Applications
- 2024.03: Β ππ Invited Talk at UT Dallas, Advancing Immersive Computing Systems in Age of Machine Learning.
- 2024.03: Β ππ Paper Accepted in ACM MobiSysβ24, NeRFHub: A Context-Aware NeRF Serving Framework for Mobile Immersive Applications.
- 2024.01: Β ππ Paper Accepted in ACM MMSysβ24, Learning to Manage Uncertainty in Video Streaming.
- 2023.12: Β ππ Paper Accepted in USENIX NSDIβ24, LiFteR: Unleash Learned Codecs in Video Streaming with Loose Frame Referencing.
π Selected Publications
- ImmerScope: Multi-view Video Aggregation at Edge towards Immersive Content Services, Bo Chen, Hongpeng Guo, Mingyuan Wu, Zhe Yang, Zhisheng Yan, Klara Nahrstedt, SenSys 2024
- LiFteR: Unleash Learned Codecs in Video Streaming with Loose Frame Referencing, Bo Chen, Zhisheng Yan, Yinjie Zhang, Zhe Yang, Klara Nahrstedt, NSDI 2024
- NeRFHub: A Context-Aware NeRF Serving Framework for Mobile Immersive Applications, Bo Chen, Zhisheng Yan, Bo Han, Klara Nahrstedt, MobiSys 2024
- Vesper: Learning to Manage Uncertainty in Video Streaming, Bo Chen, Mingyuan Wu, Hongpeng Guo, Zhisheng Yan, Klara Nahrstedt, MMSys 2024
- [Best Student Paper] Context-aware Image Compression Optimization for Visual Analytics Offloading, Bo Chen, Zhisheng Yan, Klara Nahrstedt, MMSys 2022, Media Coverage: Siebel School News, George Mason University News
- [Best Paper Award] SEAWARE: Semantic Aware View Prediction System for 360-degree Video Streaming, Jounsup Park, Mingyuan Wu, Eric Lee, Bo Chen, Klara Nahrstedt, Michael Zink, Ramesh Sitaraman, ISM 2020
- Please check the full list at Google Scholar or dblp.
π Honors and Awards
- 2022.06 Best Student Paper Award (ACM MMSys 2022).
- 2020.12 Best Paper Award (IEEE ISM 2022).
- 2019.06 SIGMM Travel Grant (ACM MMSys 2019).
π Educations
- 2022.07 - 2024.04 (now), Postdoctoral Research Associate at University of Illinois at Urbana-Champaign, Champaign, IL, USA.
- 2016.09 - 2022.05, Ph.D. at University of Illinois at Urbana-Champaign, Champaign, IL, USA.
- 2012.09 - 2016.06, B.S. at Shanghai Jiao Tong University, Shanghai, China.
- 2009.09 - 2012.06, High school at Fuzhou No.1 High School, Fuzhou, China.
π» Internships
- 2020.05 - 2020.08, Student intern at Meta.
- 2019.05 - 2019.07, Research intern at AT&T.
π Teaching
- 2023, CS 537 Advanced Topics in IOT, Teaching Assistant, University of Illinois at Urbana-Champaign.
- 2022, CS 537 Advanced Topics in IOT, Teaching Assistant, University of Illinois at Urbana-Champaign.
- 2020, CS 438 Communication Network, Teaching Assistant, University of Illinois at Urbana-Champaign.
π¬ Services
- Reviewer, 2025: ACM TOMM, IEEE TMM, 2024: ACM MM, IEEE ICCCN, ACM TOMM, 2023: ACM MM, ACM MMSys, ACM TOMM.
- TPC Member, 2025: ACM MMSys, 2024: ACM MMSys, 2023: SEC, ImmerCOM.
- Publication Chair, 2024: NSF Workshop on Sustainable Computing for Sustainability, 2023: IEEE SECON.