可流式神经视频压缩：面向跨平台部署的混合精度方法
Streamable Neural Video Compression: A Mixed Precision Approach for Cross-Platform Deployment
浏览论文内容
中文总结 AI 辅助
本文针对神经视频编解码器跨平台部署的浮点非确定性问题，提出混合精度策略，实现1080p下GPU跨代无缝解码且压缩效率损失可忽略，验证了可流式学习视频压缩的可行性。
中文摘要 AI 辅助
神经视频编解码器（Neural Video Codecs, NVC）具备前所未有的率失真性能，在5G蜂窝网络、新兴卫星直连小区（Direct-to-Cell, D2C）链路等带宽受限环境中极具应用潜力。然而，NVC在实际流式应用部署中面临跨平台浮点非确定性问题，会导致算术熵编码器在不同GPU架构间失步崩溃。近期基于整数的量化方法虽能缓解该问题，但INT8会造成压缩效率大幅下降，INT16则因绕过硬件加速引发严重计算瓶颈。本文提出一种可流式的客户端-服务器NVC架构，采用新型混合精度（FP16/FP32）策略：在硬件加速的FP16中策略性执行P帧以保障实时吞吐量，同时将I帧及周期性特征适配器重置强制为符合IEEE-754标准的FP32，确保关键边界的确定性同步。通过在涵盖四个架构代际的12款GPU上开展广泛的编解码交叉评估，证明该方法成功消除代内碎片化，大幅提升跨芯片互操作性，实现1080p分辨率下近期架构间的无缝跨代解码，且对压缩效率影响可忽略不计。此外，本文在Wi-Fi 6、5G NR（FDD/TDD）、星链D2C等多种实际网络中评估系统端到端延迟，验证了可流式学习视频压缩的实际可行性，同时凸显了非地面网络中的独特挑战。
英文摘要
Neural Video Codecs (NVCs) offer unprecedented rate-distortion performance, making them highly attractive for bandwidth-constrained environments like 5G cellular networks and emerging satellite direct-to-cell (D2C) links. However, deploying NVCs in real-world streaming applications is severely hindered by cross-platform floating-point non-determinism, which causes arithmetic entropy coders to desynchronize and crash across different GPU architectures. While recent integer-based quantization methods address this, they incur either massive degradation in compression efficiency (INT8) or severe computational bottlenecks by bypassing hardware acceleration (INT16). In this paper, we propose a streamable, client-server NVC architecture featuring a novel Mixed Precision (FP16/FP32) strategy. By strategically executing P-frames in hardware-accelerated FP16 for real-time throughput, while forcing I-frames and periodic feature-adapter resets to IEEE-754 compliant FP32, we guarantee deterministic synchronization at critical boundaries. Through extensive cross-encode/decode evaluations across 12 GPUs spanning four architectural generations, we demonstrate that our approach successfully eliminates intra-generation fragmentation and substantially broadens cross-die interoperability, achieving seamless cross-generation decodability for recent architectures at 1080p. Crucially, this is achieved with a negligible impact on compression efficiency. Furthermore, we evaluate the system's end-to-end latency across diverse real-world networks, including Wi-Fi 6, 5G NR (FDD/TDD), and Starlink D2C, proving the practical viability of streamable learned video compression while highlighting unique challenges in Non-Terrestrial Networks.
