Ahmed Elsharkawy

Mr. Ahmed Elsharkawy

Gwangju Institute of Science and Technology (GIST)
+1
orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 May 2024
Verified
Conference Paper
Open access
Ahmed ElsharkawyAhmed Elsharkawy
Ahmed Elsharkawy
AAAya Abdulnasser Saed Ataya
Aya Abdulnasser Saed Ataya
DYDohyeon Yeo
Dohyeon Yeo
EAEunsol An
Eunsol An
SHSeokhyun Hwang
Seokhyun Hwang
SeungJun KimSeungJun Kim
SeungJun Kim
Passengers can engage more in nondriving-related tasks owing to recent advancements in autonomous vehicles (AVs), making immersive tools such as virtual reality (VR) appealing; however, motion sickness (MS) remains a significant challenge. We present SYNC-VR, a system that aligns with visual, haptic, and auditory cues and provides proprioceptive feedback to illustrate its effect on MS and presence within the in-vehicle VR. We conducted an experiment with 24 participants using a real vehicle along a route with known MS-triggering events. Using subjective and physiological measures, we assessed participants’ presence and MS under four conditions by gradually varying the level of synchronized input sensations. Results reveal that SYNC-VR reduces MS and increases the sense of presence. Additionally, it emphasizes the impact of our interactive VR content and its role in achieving proprioceptive feedback with haptic feedback through electrical muscle stimulation, introducing an innovative approach to MS mitigation in in-vehicle VR.

23 December 2021 in Sensors
Verified
Journal Article
Open access
PDF file
SJsanghun jeon
sanghun jeon
Ahmed ElsharkawyAhmed Elsharkawy
Ahmed Elsharkawy
Munsang KimMunsang Kim
Munsang Kim
In visual speech recognition (VSR), speech is transcribed using only visual information to interpret tongue and teeth movements. Recently, deep learning has shown outstanding performance in VSR, with accuracy exceeding that of lipreaders on benchmark datasets. However, several problems still exist when using VSR systems. A major challenge is the distinction of words with similar pronunciation, called homophones; these lead to word ambiguity. Another technical limitation of traditional VSR systems is that visual information does not provide sufficient data for learning words such as “a”, “an”, “eight”, and “bin” because their lengths are shorter than 0.02 s. This report proposes a novel lipreading architecture that combines three different convolutional neural networks (CNNs; a 3D CNN, a densely connected 3D CNN, and a multi-layer feature fusion 3D CNN), which are followed by a two-layer bi-directional gated recurrent unit. The entire network was trained using connectionist temporal classification. The results of the standard automatic speech recognition evaluation metrics show that the proposed architecture reduced the character and word error rates of the baseline model by 5.681% and 11.282%, respectively, for the unseen-speaker dataset. Our proposed architecture exhibits improved performance even when visual ambiguity arises, thereby increasing VSR reliability for practical applications.

11 October 2024
Verified
Conference Paper
Open access
SKSeongjun Kang
Seongjun Kang
GKGwangbin Kim
Gwangbin Kim
SHSeokhyun Hwang
Seokhyun Hwang
JPJeongju Park
Jeongju Park
Ahmed ElsharkawyAhmed Elsharkawy
Ahmed Elsharkawy
SeungJun KimSeungJun Kim
SeungJun Kim
This study introduces "Flip-Pelt," a motor-driven peltier device designed to provide rapid thermal stimulation and congruent pressure feedback in virtual reality (VR) environments. Our system incorporates eight motor-driven peltier elements, allowing for the flipping of preheated or cooled elements to the opposite side. In evaluating the Flip-Pelt device, we assess user ability to distinguish between heat/cold sources by their patterns and stiffness, and its impact on enhancing haptic experiences in VR content that involves contact with various thermal sources. Our findings demonstrate that rapid thermal stimulation and congruent pressure feedback provided by Flip-Pelt enhance the recognition accuracy of thermal patterns and the stiffness of virtual objects. These features also improve haptic experiences in VR scenarios through their temporal congruency between tactile and thermal stimuli. Additionally, we discuss the scalability of the Flip-Pelt system to other body parts by proposing design prototypes.

11 May 2024
Verified
Conference Paper
Open access
SHSeokhyun Hwang
Seokhyun Hwang
JOJeongseok Oh
Jeongseok Oh
SKSeongjun Kang
Seongjun Kang
MSMinwoo Seong
Minwoo Seong
Ahmed ElsharkawyAhmed Elsharkawy
Ahmed Elsharkawy
SeungJun KimSeungJun Kim
SeungJun Kim
This study presents ErgoPulse, a system that integrates biomechanical simulation with electrical muscle stimulation (EMS) to provide kinesthetic force feedback to the lower-body in virtual reality (VR). ErgoPulse features two main parts: a biomechanical simulation part that calculates the lower-body joint torques to replicate forces from VR environments, and an EMS part that translates torques into muscle stimulations. In the first experiment, we assessed users’ ability to discern haptic force intensity and direction, and observed variations in perceived resolution based on force direction. The second experiment evaluated ErgoPulse’s ability to increase haptic force accuracy and user presence in both continuous and impulse force VR game environments. The experimental results showed that ErgoPulse’s biomechanical simulation increased the accuracy of force delivery compared to traditional EMS, enhancing the overall user presence. Furthermore, the interviews proposed improvements to the haptic experience by integrating additional stimuli such as temperature, skin stretch, and impact.

21 March 2021 in Sensors
Verified
Journal Article
Open access
PDF file
Aya AtayaAya Ataya
Aya Ataya
Won KimWon Kim
Won Kim
Ahmed ElsharkawyAhmed Elsharkawy
Ahmed Elsharkawy
SeungJun KimSeungJun Kim
SeungJun Kim
Autonomous vehicle technology increasingly allows drivers to turn their primary attention to secondary tasks (e.g., eating or working). This dramatic behavior change thus requires new input modalities to support driver–vehicle interaction, which must match the driver’s in-vehicle activities and the interaction situation. Prior studies that addressed this question did not consider how acceptance for inputs was affected by the physical and cognitive levels experienced by drivers engaged in Non-driving Related Tasks (NDRTs) or how their acceptance varies according to the interaction situation. This study investigates naturalistic interactions with a fully autonomous vehicle system in different intervention scenarios while drivers perform NDRTs. We presented an online methodology to 360 participants showing four NDRTs with different physical and cognitive engagement levels, and tested the six most common intervention scenarios (24 cases). Participants evaluated our proposed seven natural input interactions for each case: touch, voice, hand gesture, and their combinations. Results show that NDRTs influence the driver’s input interaction more than intervention scenario categories. In contrast, variation of physical load has more influence on input selection than variation of cognitive load. We also present a decision-making model of driver preferences to determine the most natural inputs and help User Experience designers better meet drivers’ needs.

23 March 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Ahmed ElsharkawyAhmed Elsharkawy
Ahmed Elsharkawy
Khawar NaheemKhawar Naheem
Khawar Naheem
DKDongwoo Koo
Dongwoo Koo
Munsang KimMunsang Kim
Munsang Kim
With the rapid development of interactive technology, creating systems that allow users to define their interactive envelope freely and provide multi-interactive modalities is important to build up an intuitive interactive space. We present an indoor interactive system where a human can customize and interact through a projected screen utilizing the surrounding surfaces. An ultra-wideband (UWB) wireless sensor network was used to assist human-centered interaction design and navigate the self-actuated projector platform. We developed a UWB-based calibration algorithm to facilitate the interaction with the customized projected screens, where a hand-held input device was designed to perform mid-air interactive functions. Sixteen participants were recruited to evaluate the system performance. A prototype level implementation was tested inside a simulated museum environment, where a self-actuated projector provides interactive explanatory content for the on-display artifacts under the user’s command. Our results depict the applicability to designate the interactive screen efficiently indoors and interact with the augmented content with reasonable accuracy and relatively low workload. Our findings also provide valuable user experience information regarding the design of mobile and projection-based augmented reality systems, with the ability to overcome the limitations of other conventional techniques.