QADRA

QADRA (Quality-Aware Dynamic Resolution Adaptation) is an open-source Python-based framework designed for adaptive video streaming applications. It determines encoding resolutions and quantization parameters to maximize perceptual quality while adhering to encoding and decoding t

Link

• Visit resource

Related resources

• DeepGameReal-time content-adaptive encoding for cloud gaming using spatiotemporal DNN to weight regions by player gaze/interest, cutting bandwidth…
• Sustainable Real-Time 8K60 HEVC Encoding for V2XPaper on repurposing legacy Pascal-generation NVENC hardware for real-time 8K60 HEVC encoding via split-frame encoding, targeting V2X appli…
• Comparing H.265/HEVC and VP9: Impact of High Frame Rates on Perceptual QualityAcademic comparison of HEVC and VP9 codecs under high frame-rate conditions, including ffmpeg/libvpx tuning parameters used.
• FPGA-based Logarithmic Hop Encoding for Low-Latency Real-Time Video (Drones)Peer-reviewed paper on an FPGA-implemented low-latency real-time video encoder (LHE) for embedded/drone applications, with companion code r…
• Cool-chic: Coordinate-based Low Complexity Hierarchical Image/Video CodecUltra-low-complexity overfitted/coordinate-based learned video codec using ~800 parameters, with companion open-source implementation.
• FOSDEM 2024: VVdeC<>Arm — Optimizing an Open Source VVC Decoder for Arm ArchitecturesFOSDEM Open Media devroom talk detailing Arm-specific performance optimization work on the open-source VVdeC VVC decoder.
• nano-hevcMinimal educational HEVC encoder written in pure Python, readable-over-fast, for teaching codec internals.
• Differentiable VMAF Re-implementation on PyTorchDifferentiable PyTorch reimplementation of VMAF metric enabling gradient-based codec/RDO optimization, with companion code.
• Performance Comparison of AV1, JEM, VP9, and HEVC EncodersFraunhofer HHI SPIE 2017 paper comparing codec internals, BD-BR, and complexity across AV1, JEM, VP9, and HEVC.
• Opus (xiph/opus)Reference implementation of the Opus audio codec, recently adding deep-learning-based packet loss concealment.