Embedded Files
Our research spans systems and AI, including (but not limited to):
Computer and Networked Systems
AI-driven Multimedia Processing and Delivery
Systems for AI / AI for Systems
Immersive Computing (e.g., XR Systems and Digital Twins)
We ask fundamental questions about modern computing and networked systems:
How can we make (AI-driven) systems faster and more responsive?
Can we achieve better performance without increasing cost or energy?
Can we prioritize what matters most and adapt to dynamic workloads?
How can systems collaborate and scale efficiently?
How can AI improve real-world systems, not just models?
Below are selected demo videos and images showcasing our research.
Live Volumetric Video Processing and Streaming
Live Volumetric Video Processing and Streaming
Related Papers:
[C] BlenDR: Bandwidth-efficient RGB-D Representation and Delivery for Live 3D Video Streaming [MobiSys'26]
Large-scale 3D/4D Real-world Reconstruction Systems
Related Papers:
[J] SceneHub4D: A Dataset and Evaluation Framework for 6-DoF 4D VR Scenes [IEEE TVCG 2026, VR 2026]
[W] Reconstructing Reality over Time: From Drone Capture to Timelapse Gaussian Splatting
AI-augmented Internet Video Streaming
Related Papers:
[C] NeuroScaler: Neural Video Enhancement at Scale [SIGCOMM'22]
[C] Neural-Enhanced Live Streaming: Improving Live Video Ingest via Online Learning [SIGCOMM'20]
[C] Neural Adaptive Content-Aware Internet Video Delivery [OSDI'18]
Intelligent Networked Systems
Related Papers:
[C] OutRAN: Co-Optimizing for Flow Completion Time in Radio Access Network [CoNEXT'22]
[C] FlexPass: A Case for Flexible Credit-Based Transport for Datacenter Networks [EuroSys'23]
AI-native Media Processing and System
AI-native Media Processing and System
Related Papers:
[C] Neural-Centric Video Processing Pipeline for Unified Multi-Task Inference [CVPR'26]
[C] NerVast: Compression-Efficient Scaling of Implicit Neural Video Representations via Scene-based Parameter-sharing [WACV'26]
[W] Presto: Hybrid CPU-GPU Preprocessing Framework for Video-based AI Inference System
[J] Efficient Disaggregated Cloud Storage for Cold Videos with Neural Enhancement [IEEE Micro'26]
Page updated
Google Sites
Report abuse