Scaling bitrate in video compression using tuned parameters for efficient video analytics

Efficient Video Analytics is crucial for applications such as traffic monitoring, smart cities, and intelligent transportation systems. This study introduces a dynamic bitrate optimization approach that fine-tunes video compression parameters, including Frame Per Second (FPS), Resolution, Quantization Parameter (QP), Entropy Coding, Motion Estimation, and Macroblock Partitioning. The goal is to strike a balance between video quality, computational complexity, and bandwidth efficiency, particularly in resource-constrained environments. Experiments on Pedestrian and Road Traffic video datasets demonstrate that the proposed approach effectively reduces bandwidth consumption while maintaining Video Analytics accuracy. The results show that optimizing FPS and QP is crucial to achieving the best trade-off between efficiency and performance. Furthermore, analysis of F1-Score and Bitrate trade-offs reveals that lower-bitrate videos can maintain stable accuracy with minimal bandwidth adjustments, making them ideal for constrained network conditions. Furthermore, empirical profiling indicates that Context Adaptive Binary Arithmetic Coding (CABAC), lightweight Motion Estimation, and medium-grain Macroblock Partitioning achieve practical compromises by reducing runtime overhead while preserving accuracy. These findings highlight the importance of adaptive compression strategies for enabling scalable, sustainable Video Analytics. By dynamically adjusting compression parameters, Video Analytics systems can operate more efficiently in smart city infrastructure, ensuring reliable performance while reducing resource consumption. • Proposed a dynamic framework for tuning video compression parameters to optimize bitrate. • The method was evaluated using Pedestrian and Road Traffic video datasets and performance metrics. • Achieved significant bandwidth reduction while maintaining the accuracy of the analysis. • Provides scalable solutions for smart cities and intelligent transportation systems. • Demonstrates the integration of compression and resource allocation for efficiency.