Title: Deep learning-enhanced residual coding techniques for specialised video streaming applications

Authors: Shilpa Bagade; Jyothi Sri Vadlamudi; Kiran Mannem; Jamal Kovelakuntla

Addresses: Department of Information Technology, School of Engineering, Malla Reddy University, Hyderabad, India ' Department of Electronics and Engineering, Malla Reddy (Deemed to be University), Hyderabad, India ' Department of Electronics and Communication Engineering, Gokaraju Rangaraju Institute of Engineering and Technology, Affiliated to JNTUH, Hyderabad, India ' Department of Electronics and Communication Engineering, Gokaraju Rangaraju Institute of Engineering and Technology, Affiliated to JNTUH, Hyderabad, India

Abstract: An increasing demand for superior and effective video streaming has made developments in residual coding crucial. This work explores the improvement of residual coding methods by employing deep learning for specific video streaming applications, with a particular emphasis on the H.265 (HEVC) standard. Using advanced deep learning techniques, an enhanced residual coding system is introduced that greatly enhances compression efficiency and video quality. Extensive experiments are conducted to compare the deep learning-enhanced H.265 framework with the conventional H.264 standard. The results demonstrate significant enhancements in important parameters such as peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). The approach not only achieves a better level of compression for video files but also preserves a greater level of visual quality, demonstrating its potential to greatly improve video streaming applications that are particular to certain fields or industries.

Keywords: deep learning; residual coding; video streaming; HEVC; peak signal-to-noise ratio; PSNR; structural similarity index measure; SSIM.

DOI: 10.1504/IJSCC.2026.155225

International Journal of Systems, Control and Communications, 2026 Vol.17 No.3, pp.269 - 280

Received: 07 Dec 2024
Accepted: 12 May 2025
Published online: 29 Jul 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article