1. Efficient Image and Video Coding for Multimedia Communications

This research direction advances efficient video coding methods to support high-quality multimedia communications under bandwidth and computation constraints. Contributions span classical HEVC/SHVC scalable coding—where multiple studies improved prediction structures, merge modes, side-information generation, and complexity control—to modern learning-based video compression frameworks. Recent works introduce neural B-frame coding, wavelet-driven image compression, and perceptual rate–quality modeling, tackling challenges such as domain shift and quality consistency. Complementary research on enhancement networks and edge-guided super-resolution enables robust video delivery for surveillance and industrial multimedia systems. Together, these contributions significantly improve compression efficiency, visual quality, and adaptability, supporting real-time multimedia communication across diverse platforms.

Learned Video Coding / Neural Video Compression (2023–2025):
HEVC/SHVC/Scalable Video Coding (2011–2018):

(Selected publication) Adaptive Scalable Video Coding: An HEVC-Based Framework Combining the Predictive and Distributed Paradigms, ( IEEE Transactions on Circuits and Systems for Video Technology, Volume: 27, Issue: 8, August 2017)

Abstract:

The emerging scalable High Efficiency Video Coding (SHVC) video coding standard provides an efficient solution for transmission of video over heterogeneous and time dynamic networks, terminals, and usage environments. The encoding complexity and the error sensitivity associated with the efficient HEVC coding tools adopted in SHVC make this scalable codec less attractive to some emerging applications such as video surveillance, visual sensor network, and remote space transmission where these requirements are critical. To address the requirements of these application scenarios including scalability, this paper proposes a novel HEVC-based framework offering quality scalability on top of an HEVC compliant base layer while appropriately combining the predictive and distributed coding paradigms. To achieve the best enhancement layer compression efficiency, two novel coding tools are proposed, notably a machine learning-based side information creation mechanism and an adaptive correlation modeling process. The experimental results reveal that the rate-distortion performance of the proposed distributed scalable video coding-HEVC solution outperforms the relevant alternative coding solutions, notably by up to 52.9% and 23.7% BD-rate gains regarding the HEVC-Simulcast and SHVC standard solutions, respectively, for an equivalent prediction configuration, while achieving a lower encoding complexity.

Joint Layer Prediction for Improving SHVC Compression Performance and Error Concealment (IEEE Transactions on Broadcasting ( Volume: 65, Issue: 3, September 2019)

Abstract:

Scalable high efficiency video coding (SHVC) standard is expected to play a more important role in the heterogeneous landscape of broadcasting, multimedia, networks, and various services applications as it is specified as a layered coding technique in the advanced television systems committee 3.0. However, its block-based structure of temporal and spatial prediction makes it sensitive to information loss and error propagation due to transmission errors. In this context, we propose an improved SHVC with a joint layer prediction (JLP) solution which adaptively combines the decoded information from the base and the enhancement layers to create an additional reference for the SHVC enhancement encoder. To optimize the quality of the joint prediction, the minimum mean square error estimation is executed in computing a combination factor which gives weights to each contribution of the decoded information from the layers. In addition, the proposed JLP is integrated into the SHVC decoder to work as an error concealment solution to mitigate the error propagation happening inevitably in practical video transmission. Experiments have shown that the proposed SHVC framework significantly outperforms its relevant benchmarks, notably by up to 14.8% in bitrate reduction with respect to the standard SHVC codec. The proposed SHVC error concealment strategy also greatly improves the concealed picture quality as well as reducing the problem of error propagation when compared to conventional error concealment approaches.

2. Vision Processing and Understanding for Robotics

This research theme focuses on developing advanced vision-based analytics to enhance Human–Robot Interaction (HRI). Core contributions include deep-learning pipelines for human intent recognition, behavior understanding, and real-time perceptual processing to support safe and natural collaboration. Context-aware interaction frameworks leveraging large language models extend the robot’s ability to interpret user commands and situational cues. Motion-resolution adaptation algorithms ensure responsive and intuitive robot behaviors during cooperative tasks. Additional developments in humanoid and quadruped robot platforms provide experimental testbeds for embodied interaction. Collectively, these works strengthen perceptual intelligence and adaptive decision-making in robots, enabling more seamless and socially aware interactions with humans.

In addition, in this research also advances semantic SLAM and autonomous navigation through robust perception, mapping, and localization techniques. Key contributions include semantic SLAM systems that fuse geometric and contextual cues, illumination-robust fusion pipelines, and sparse semantic mapping frameworks for efficient long-term operation. Recent work integrates 3D Gaussian Splatting to enhance map fidelity and support collaborative SLAM scenarios. Complementary studies on monocular 3D object localization and motion prediction enable robots to understand dynamic environments and navigate safely among humans. Collectively, these contributions significantly improve the accuracy, robustness, and semantic richness of robot navigation systems, forming a foundation for next-generation autonomous and collaborative robots.

Semantic Visual SLAM
3D Object Localization (monocular, refined, 2D-to-3D)

Efficient Human–Robot Interaction via Deep Perception
Video enhancement & industrial video analytics:

(Selected publication) ESRPCB: An edge guided super-Resolution model and ensemble learning for tiny Printed Circuit Board defect detection, Engineering Applications of Artificial Intelligence, Volume 159, Part A, 1 November 2025, 111547

Abstract

Printed Circuit Boards (PCBs) are critical components in modern electronics, which require stringent quality control to ensure proper functionality. However, the detection of defects in small-scale PCBs images poses significant challenges as a result of the low resolution of the captured images, leading to potential confusion between defects and noise. To overcome these challenges, this paper proposes a novel framework, named ESRPCB (edge-guided super-resolution for PCBs defect detection), which combines edge-guided super-resolution with ensemble learning to enhance PCBs defect detection. The framework leverages the edge information to guide the EDSR (Enhanced Deep Super-Resolution) model with a novel ResCat (Residual Concatenation) structure, enabling it to reconstruct high-resolution images from small PCBs inputs. By incorporating edge features, the super-resolution process preserves critical structural details, ensuring that tiny defects remain distinguishable in the enhanced image. Following this, a multi-modal defect detection model employs ensemble learning to analyze the super-resolved image, improving the accuracy of defect identification. Experimental results demonstrate that ESRPCB achieves superior performance compared to State-of-the-Art (SOTA) methods, achieving an average Peak Signal to Noise Ratio (PSNR) of 30.54 , surpassing EDSR by . In defect detection, ESRPCB achieves a mAP50(mean average precision at an Intersection over Union threshold of 0.50) of 0.965, surpassing EDSR (0.905) and traditional super-resolution models by over 5%. Furthermore, the ensemble-based detection approach further enhances performance, achieving a mAP50 of 0.977. These results highlight the effectiveness of ESRPCB in enhancing both image quality and defect detection accuracy, particularly in challenging low-resolution scenarios.

Optimal design and fabrication of frame structure for dual-arm service robots: An effective approach for human–robot interaction, Engineering Science and Technology, an International Journal, Volume 56, August 2024, 101763

Abstract

Rapid advancement in robotics technology has paved the way for developing mobile service robots capable of human interaction and assistance. In this paper, we propose a comprehensive approach to design, fabricate, and optimize the overall structure of a dual-arm service robot. The conceptual design phase focuses on both critical components, the mobile platform and the manipulation system, essential for seamless navigation and effective task execution. In the proposed system, the distribution of the robot payload in terms of region, maximum stress, and displacement is examined, comprehensively analyzed, and compared with the relevant works. In addition, to enhance the system’s efficiency while minimizing its weight, we introduce a lightweight design approach in which Finite Element Analysis is utilized to optimize the frame structure. Subsequently, we fabricate a physical prototype based on the derived model. Finally, we provide a kinematic model for our dual-arm service robot and demonstrate its efficacy in both control and human–robot interaction (HRI) tasks. Experimental results indicate that the proposed dual arm design can achieve a significant weight reduction of 25% from the original design while still performing actions smoothly for HRI tasks.