Development of Duplex Eye Contact Framework for Human-Robot Inter Communication

Sharmin, Shayla; Hoque, Mohammed Moshiul; Islam, S. M. Riazul; Kader, Md. Fazlul; Sarker, Iqbal H.

doi:10.1109/access.2021.3071129

Cited by 6 publications

(3 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(e) Robot head [53], (f) Based on Xilinx Virtex-6 FPGA [52], (g) Humanoid robot neck [54], (h) Robotic head prototype with hardware configurations [55] Service Robotic Platforms. (i) Photographer Robot "Fotor" [56], (j) Based on Jetson TX1 [57], (k) Omnidirectional Mobile Robot [58]. Not all parts could be drilled or bonded, which reduced the system's modularity.…”

Section: Section Vi: Results and Discussionmentioning

confidence: 99%

See 1 more Smart Citation

The Face Detection / Recognition , Perspective and Obstacles In Robotic: A Review

Uddin

Moshayedi

et al. 2022

EAI Endorsed Trans AI Robotics

View full text Add to dashboard Cite

Facial recognition research is one of the different types of research in this world today. In recent years, facial recognition in robots has attracted increased study interest. Robotic platforms now utilize a variety of object detection methods, with face detection being a viable use. Face detection in robotics is a computer technique that recognizes human faces in digital pictures and is used in a range of applications. Different authors have performed their research in different ways on the use of detection systems. This paper aims to give future researchers a better idea of using facial recognition systems in robotics. In this study, we reviewed research by various authors over recent years to facilitate future facial recognition research. In addition, scholars have addressed the topics, how they have done so, and the specifics of their approaches are described. This paper reviewed an overview of hardware implementation and software implementation by various authors. It can automatically focus cameras or count the number of people who have entered a location. Commercial applications of the method include displaying tailored advertisements in response to a recognized face along with the algorithms, functions and architectures used in facial recognition and giving the opinions of various authors mentioned. The comparative analysis of facial recognition and its architecture system is highlighted.

show abstract

Section: Section Vi: Results and Discussionmentioning

confidence: 99%

“…The outcomes of the robotic framework in a scenario using the suggested duplex eye contact mechanism are reported. The findings reveal that, for the human initiative case and the robot initiative example, the suggested strategy made eye contact with 92% and 86% accuracy [56].…”

Section: P B Nithin Et Al (mentioning

confidence: 94%

The Face Detection / Recognition , Perspective and Obstacles In Robotic: A Review

Uddin

Moshayedi

et al. 2022

EAI Endorsed Trans AI Robotics

View full text Add to dashboard Cite

show abstract

“…Semantic segmentation [1], [2], [3] is a fundamental computer vision task that aims to classify each pixel of the given image and assign the corresponding class label. Despite the remarkable achievement in the traditional semantic segmentation field where samples for all classes are available, continuous learning of data streams in real-world scenarios [4], [5] remains challenging. In recent years, researchers are widely interested in Class-incremental Semantic Segmentation (CISS), which means the semantic segmentation model has to learn newly emerged classes incrementally while preserving its capabilities of segmenting object(s) of old classes without any old examples.…”

Section: Introductionmentioning

confidence: 99%

Class-Prompting Transformer for Incremental Semantic Segmentation

Song,

Shi,

Shang

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Class-incremental Semantic Segmentation (CISS) aims to learn new tasks sequentially that assign a specific category to each pixel of a given image while preserving the original capability to segment the old classes even if the labels of old tasks are absent. Most existing CISS methods suppress catastrophic forgetting by directly distilling on specific layers, which ignores the semantic gap between training data from the old and new classes with different distributions and leads to distillation errors, thus affecting segmentation performance. In this paper, we propose a Class-prompting Transformer (CPT) to introduce external prior knowledge provided by a pre-trained vision-language encoder into CISS pipelines for bridging the old and new classes and performing more generalized initialization and distillation. Specifically, we proposed a Prompt-guided Initialization Module (PIM), which measures the relationships between the class prompts and old query parameters to initialize the new query parameters for relocating the previous knowledge to the learning of new tasks. Then, a Semantic-aligned Distillation Module (SDM) is proposed to incorporate class prompt information with the class-aware embeddings extracted from the decoder to prevent the semantic gap problem between distinct class data and conduct adaptive knowledge transfer to suppress catastrophic forgetting. Extensive experiments on Pascal VOC and ADE20K datasets for the CISS task demonstrate the superiority of the proposed method, which achieves state-of-the-art performance.INDEX TERMS Incremental semantic segmentation, knowledge distillation, class prompt learning.LINFENG XU (Member, IEEE) received the Ph.D. degree in signal and information processing from the School of Electronic Engineering,

show abstract