Awesome Media Delivery  [draft]

Papers User-Oriented ABR SODA: An Adaptive Bitrate Controller for Consistent High-Quality Video Streaming [SIGCOMM 2024] Veritas: Answering Causal Queries from Video Streaming Traces [SIGCOMM 2023] Sammy: smoothing video traffic to be a friendly internet neighbor [SIGCOMM 2023] Genet: Automatic Curriculum Generation for Learning Adaptation in Networking [SIGCOMM 2022] ARTEMIS: Adaptive Bitrate Ladder Optimization for Live Video Streaming [NSDI 2024] Learning in situ: a randomized experiment in video streaming [NSDI 2020] Rosevin: Employing Resource- and Rate-Adaptive Edge Super-Resolution for Video Streaming [INFOCOM 2024] Meta Reinforcement Learning for Rate Adaptation [INFOCOM 2023] RDladder: Resolution-Duration Ladder for VBR-encoded Videos via Imitation Learning [INFOCOM 2023] Lumos: towards Better Video Streaming QoE through Accurate Throughput Prediction [INFOCOM 2022] Adaptive Bitrate with User-level QoE Preference for Video Streaming [INFOCOM 2022] High-Volume Video NeRFHub: A Context-Aware NeRF Serving Framework for Mobile Immersive Applications [MOBISYS 2024] Dragonfly: Higher Perceptual Quality For Continuous 360° Video Playback [SIGCOMM 2023] Ekho: Synchronizing cloud gaming media across multiple endpoints [SIGCOMM 2023] ZGaming: Zero-Latency 3D Cloud Gaming by Image Prediction [SIGCOMM 2023] Pudica: Toward Near-Zero Queuing Delay in Congestion Control for Cloud Gaming [NSDI 2024] AUGUR: Practical Mobile Multipath Transport Service for Low Tail Latency in Real-Time Streaming [NSDI 2024] Habitus: Boosting Mobile Immersive Content Delivery through Full-body Pose Tracking and Multipath Networking [NSDI 2024] YuZu: Neural-Enhanced Volumetric Video Streaming [NSDI 2022] MuV2: Scaling up Multi-user Mobile Volumetric Video Streaming via Content Hybridization and Sharing [MOBICOM 2024] MetaStream: Live Volumetric Content Capture, Creation, Delivery, and Rendering in Real Time [MOBICOM 2023] AccuMO: Accuracy-Centric Multitask Offloading in Edge-Assisted Mobile Augmented Reality [MOBICOM 2023] Vues: practical mobile volumetric video streaming through multiview transcoding [MOBICOM 2022] SalientVR: saliency-driven mobile 360-degree video streaming with gaze information [MOBICOM 2022] Flexible high-resolution object detection on edge devices with tunable latency [MOBICOM 2021] ViVo: visibility-aware mobile volumetric video streaming [MOBICOM 2020] GROOT: a real-time streaming system of high-fidelity volumetric videos [MOBICOM 2020] TBSR: Tile-Based 360° Video Streaming with Super-Resolution on Commodity Mobile Devices [INFOCOM 2024] Energy-Efficient 360-Degree Video Streaming on Multicore-Based Mobile Devices [INFOCOM 2023] OmniSense: Towards Edge-Assisted Online Analytics for 360-Degree Videos [INFOCOM 2023] Collaborative Streaming and Super Resolution Adaptation for Mobile Immersive Videos [INFOCOM 2023] Robust 360° Video Streaming via Non-Linear Sampling [INFOCOM 2021] Motion-Prediction-based Wireless Scheduling for Multi-User Panoramic Video Streaming [INFOCOM 2021] Popularity-Aware 360-Degree Video Streaming [INFOCOM 2021] ARISE: High-Capacity AR Offloading Inference Serving via Proactive Scheduling [MOBISYS 2024] Theia: Gaze-driven and Perception-aware Volumetric Content Delivery for Mixed Reality Headsets [MOBISYS 2024] OmniLive: Super-Resolution Enhanced 360° Video Live Streaming for Mobile Devices [MOBISYS 2023] AdaptSLAM: Edge-Assisted Adaptive SLAM with Resource Constraints via Uncertainty Minimization [INFOCOM 2023] VR Viewport Pose Model for Quantifying and Exploiting Frame Correlations [INFOCOM 2022] Map++: Towards User-Participatory Visual SLAM Systems with Efficient Map Expansion and Sharing [MOBICOM 2024] Classic Converge: QoE-driven Multipath Video Conferencing over WebRTC [SIGCOMM 2023] CellFusion: Multipath Vehicle-to-Cloud Video Streaming with Network Coding in the Wild [SIGCOMM 2023] XRON: A Hybrid Elastic Cloud Overlay Network for Video Conferencing at Planetary Scale [SIGCOMM 2023] LiveNet: a low-latency video transport network for large-scale live streaming [SIGCOMM 2022] GSO-simulcast: global stream orchestration in simulcast video conferencing systems [SIGCOMM 2022] XLINK: QoE-driven multi-path QUIC transport in large-scale video services [SIGCOMM 2021] Hairpin: Rethinking Packet Loss Recovery in Edge-based Interactive Video Streaming [NSDI 2024] Tambur: Efficient loss recovery for videoconferencing via streaming codes [NSDI 2023] Enabling High Quality Real-Time Communications with Adaptive Frame-Rate [NSDI 2023] SENSEI: Aligning Video Streaming Quality with Dynamic User Sensitivity [NSDI 2021] Venus: Enhancing QoE of Crowdsourced Live Video Streaming by Exploiting Multiflow Viewer Assistance [MOBICOM 2024] Chorus: Coordinating Mobile Multipath Scheduling and Adaptive Video Streaming [MOBICOM 2024] Elf: accelerate high-resolution mobile deep vision with content-aware parallel offloading [MOBICOM 2021] OnRL: improving mobile video telephony via online reinforcement learning [MOBICOM 2020] AggDeliv: Aggregating Multiple Wireless Links for Efficient Mobile Live Video Delivery [INFOCOM 2024] TileSR: Accelerate On-Device Super-Resolution with Parallel Offloading in Tile Granularity [INFOCOM 2024] Who is the Rising Star?...

September 8, 2024 · 8 min · 1667 words · Yuheng

the way to master rust  [draft]

Rust I use rust version 1.76.0 (07dca489a 2024-02-04). useful links: the book rust by examples Rustonomicon rust RFC Ownership The rules of ownership: Each value in Rust has an owner There can only be one owner at a time (may have shared ownership via Arc/Rc, invoking clone produces a new pointer to the same allocation in the heap) When the owner goes out of the scope, the value will be dropped References & Borrowing We call the action of creating a reference - borrowing, which won’t take the ownership of the variable....

March 13, 2024 · 10 min · 2063 words · Yuheng

World Models  [draft]

World Models Learning the transition of the environment Reference: world models paper world models author's blog The goal of World Model is to learn a transition model which is able to characterize the probability distribution $p(s_{t}|s_{t-1}, a_{t-1})$. Such pre-knowledge of the environment transition convert the tasks into model based, thereby enabling the application of advanced methodologies, including planning and model-based reinforcement learning (MBRL). In this paper, the World Model is built with two distinct yet interconnected components: Vision Model (V) and Memory Model (M) where V learns to encode the high-dimensional observation into low-dimensional latent space and M learns to predict low-dimensional representation of the future observation....

November 29, 2023 · 9 min · 1898 words · Yuheng