Optimizing real-time stereo image retargeting for AR/VR: Lightweight disparity CNNs on AI-Driven edge architectures.
Saved in:
| Title: | Optimizing real-time stereo image retargeting for AR/VR: Lightweight disparity CNNs on AI-Driven edge architectures. |
|---|---|
| Authors: | T. Jagtap, Mahendra1,2 (AUTHOR) mtjagtap05@gmail.com, Unhelkar, Bhuvan3 (AUTHOR), Kshirsagar, Pravin R.4 (AUTHOR), Rakesh, Nitin5 (AUTHOR), Thiagarajan, R.6 (AUTHOR), Patil, Vishal7 (AUTHOR) |
| Source: | African Journal of Science, Technology, Innovation & Development. Feb2026, Vol. 18 Issue 1, p46-66. 21p. |
| Subjects: | Stereo image processing, Convolutional neural networks, Augmented reality, Depth maps (Digital image processing), Virtual reality, Computer performance, Edge computing |
| Abstract: | Real-time stereo image retargeting for augmented reality (AR) and virtual reality (VR) necessitates precise per-pixel depth estimation and ultra-low latency performance on resource-limited edge devices. Current disparity convolutional neural networks (CNNs) and retargeting pipelines are unable to meet these demanding requirements concurrently. This paper presents EASNet, a compact end-to-end framework that unifies geometry-aware proposal generation, parallax-aligned feature encoding, sparse candidate aggregation, and uncertainty-guided refinement to enable high-fidelity stereo retargeting on edge architectures. This system enhances stereo vision through Epipolar-Adaptive Disparity Proposals (EADP) for search space reduction, a Parallax-Directed Deformable Encoder (PaDDE) for improved matching in repetitive and low-texture areas, Sparse Epipolar Candidate Volume (SECV) with Edge-Consistent Routing (ECR) for efficient, boundary-preserving cost aggregation, and Lightweight Uncertainty-Guided Refinement (LUGR) for sub-pixel structure and occlusion correction. Evaluated on high-resolution indoor stereo data, EASNet attains a favourable trade-off between accuracy and efficiency (≈0.21 M parameters, 3.42 GFLOPs) while improving disparity fidelity and visual coherence required for retargeting (reported EPE ≈ 1.78 px, D1 ≈ 5.02%, VC ≈ 93.7%). The design emphasizes quantization compatibility and deterministic latency, enabling practical deployment on AR/VR edge devices. We analyze ablations, per-scene behaviour, and k-fold stability, discussing limitations like indoor bias, extreme occlusion, large baselines, and future EASNet extensions. [ABSTRACT FROM AUTHOR] |
| Copyright of African Journal of Science, Technology, Innovation & Development is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
Be the first to leave a comment!
