Day 2 :
Keynote Forum
Soo yeung lee
Korea Advanced Institute for Science and Technology, Republic of Korea
Keynote: Digital Companion with Human-like Emotion and Ethics
Time : 09:30-10:15
Biography:
Soo-Young Lee has worked on the artificial cognitive systems with human-like intelligent behavior based on the biological brain information processing. His research includes speech and image recognition, natural language processing, situation awareness, top-down attention, internal-state recognition and human-like dialog systems. Especially, among many internal states, he is interested in emotion, sympathy, trust and personality. He led Korean Brain Neuroinformatics Research Program from 1998 to 2008 with dual goals, i.e., understanding brain information processing mechanism and developing intelligent machine based on the algorithm. He is currently the Director of KAIST Institute for Artificial Intelligence and leading Emotional Conversational Agent Project and a Korean National Flagship AI Project. He was the President of Asia-Pacific Neural Network Society in 2017 and had received Presidential Award from INNS and Outstanding Achievement Award from APNNA.
Abstract:
For the successful interaction between human and digital companion, i.e., machine agents, the digital companions need be able to bind with human-like ethics as well as to make emotional dialogue, understand human emotion and express its own emotion. In this talk we present our approaches to develop human-like ethical and emotional conversational agents as a part of the Korean Flagship AI Program. The emotion of human users is estimated from text, audio and visual face expression during verbal conversation and the emotion of intelligent agents is expressed by speech and facial expression. Specifically, we will show how our ensemble networks won the Emotion Recognition in the Wild (EmotiW2015) challenge with 61.6% accuracy to recognize seven emotions from facial expression. Then, a multimodal classifier combines text, voice and facial video for better accuracy. Also, a deep learning based Text-to-Speech (TTS) system will be introduced to express emotions. These emotions of human users and agents interacts each other during the dialogue. Our conversational agents have chitchat and Question-and-Answer (Q&A) modes and the agents respond differently for different emotional states in chitchat mode. Then, the internal states will be further extended into trustworthiness, implicit intention and personality. Also, we will discuss how the agents may learn human-like ethics during the human-machine interactions.
Keynote Forum
Mingcong Deng
Tokyo University of Agriculture and Technology JAPAN
Keynote: Operator-based nonlinear control of micro hand and its application
Time : 10:15 -11:00
Biography:
Abstract:
Keynote Forum
Qingsong Xu
University of Macau, China
Keynote: Design and Application of Force-Sensing Robotic Bio-Micromanipulation Systems
Time : 11:15-12:00
Biography:
Qingsong Xu is the Director of Smart and Micro/Nano Systems Laboratory and Associate Professor of Electromechanical Engineering at the University of Macau. He was a Visiting Scholar at the University of California, Los Angeles (UCLA), USA in 2016, the RMIT University, Melbourne, Australia in 2016, the National University of Singapore, Singapore in 2012 and the Swiss Federal Institute of Technology (ETH Zurich), Switzerland in 2011. His current research area involves micro/nano-mechatronics and systems, control and automation and applications of computational intelligence. He is a Senior Member of IEEE and a Technical Editor of IEEE/ASME Transactions on Mechatronics. He has published three monographs and over 240 technical papers in international journals and conferences.
Abstract:
Robotic micromanipulation systems are demanding devices to realize automated manipulation of biological samples. Majority of existing robotic bio-micromanipulation systems work based on displacement sensing and control. The lack of force sensing prevents the wide application of the devices. In modern biotech industry, there are increasing needs for advanced micromanipulation equipment with microforce sensing and control capabilities. The development of force-sensing microinjectors and microgrippers enable extensive applications involving biological field with guaranteed safety and accuracy of advanced robotic manipulation. This talk reports our recent work on design and development of new force-sensing robotic micromanipulation systems dedicated to biological micromanipulation tasks. In particular, force-sensing microinjector and force-sensing microgripper are presented as typical examples. New microforce sensor design is conducted in details. Novel control schemes have been developed to fuse the position and force control to enable a safe and reliable manipulation. The effectiveness of the systems has been demonstrated by carrying out microinjection and microgripping operation of biological cells. The developed force-sensing robotic bio-micromanipulation systems have demonstrated wide applications in the fields of biomedical engineering, gene engineering and so on.
Keynote Forum
Manchita Dumlao
The Philippine Women’s University, Philippines
Keynote: Central emergency response management system
Time : 12:00-12:45
Biography:
Menchita F Dumlao is currently the Research Director of Philippine Women’s University, Philippines. Concurrently, she is an Associate Professor and Program Chair of Department of Information Technology and Technology Consultant of Imergex Information Technology, Inc. Her research interest is in the area of data science, machine learning and artificial intelligence.
Abstract:
Central Emergency Response Management system (CERM) facilitates estimation of dispatch response time and route or direction from the origin of the resource to incident site. The dispatch determined the capability required to a specific incident. Its core technology is Call Taking Management System (CTMS) which applies stochastic optimization to collaborate sub-systems and make everything else complete. They are connected end to end to create a consolidated database system leading to a criminal information system and incident command system which includes call taking and logging module, resource unit dispatch and geo-mapping module, SMS messaging services module, resource availability and dispatch. It can record audio patch from the telephone line through the audio input of the PC (client computer). The provided information on incident, incident type, location and other important information needed by the dispatcher (police). The map provides an aerial view (satellite image) of the location of an incident for the nearest equivalent search parameter within 10,000 meters from the center of crime. This technology shows the best route possible from the unit resource (of the crime) origin to the incident site. The control also shows text-based instruction, estimated time and distance.
- Special Session
Location: Polaris II
Session Introduction
Dr. Jun Kurihara
The Canon Institute for Global Studies, Japan
Title: The significance of robot safety standards for the development of life support robots
Time : 15:15-16:15
Biography:
Jun Kurihara is a Research Director of The Canon Institute for Global Studies (CIGS), a Tokyo-based think tank (2009 to till date). He also serves as a Corporate Director of a Japanese company (Ono Pharmaceutical, 2013 to till date) and teaches as a Visiting Professor at Kwansei Gakuin University located in Hyogo Prefecture, Japan (2006 to till date). Between 2003 and 2012, he was a Senior Fellow at the Harvard Kennedy School, Ash Center and the Center of Business and Government (CBG). Currently he is working on Japanese companies’ innovation strategies including service robotics and artificial intelligence. He has served as a Chairman and Member of various committees established by the government and business organizations including a Senior Member of the Japan Statistics Council for the Government of Japan. He has earned an MS from the Graduate School Agriculture of Kyoto University.
Abstract:
Japan is the forerunner in the world in terms of aging population. Under these circumstances, life support robots are promising tools to lead to a higher Quality Of Life (QOL) for the elderly along with physically challenged people; both young and old. However, there has been a jumble of ideas and prototypes, leading to a cul-de-sac with respect to the future development of caregiver and self-reliance support robots. This presentation proposes a promising approach to this cul-de-sac. This impasse has been caused by three factors. First, robots are extremely expensive, unaffordable for caregivers and patients. Second, robot markets are extremely compartmentalized and isolated. Third, the primary cause of compartmentalized market has been a confusing patchwork of standards. Thus far, the cul-de-sac has brought about a lackadaisical growth of life support robots, compared with the case of industrial robots that are experiencing a galloping growth, especially in East Asia. Accordingly, the time has come to draw up a new road map toward a higher speed of robot diffusion. The most effective approach would be the establishment of globally encompassing and trustworthy safety standards. They could provide a firm foundation for a globalized and integrated market of life support robots. Individual markets, thanks to universally applicable safety standards, would be loosely interconnected; those markets would include not only the market for elderly care, but also medical, educational, business and military services as well as industrial robots, leading to a larger pool of components and related technologies. The resulting larger pool could reduce the prices of life support robots and accelerate their diffusion. The establishment of globally encompassing safety standards requires an institutional framework which could play a leadership role. Japan has a robot safety center (RSC); only one in the world to systematically propose safety standards and guidelines for life support robots.
- Poster Presentations
Location: Polaris II
Chair
Emdad Khan
Internet Speech, USA
Session Introduction
Baoming Zhang
Zhengzhou University, China
Title: Deep feature learning for unsupervised change detection in high-resolution multi-temporal and multi-source images
Time : 16:15-16:35
Biography:
Baoming Zhang is currently a full Professor at Department of Photogrammetry and Remote Sensing, Zhengzhou Institute of Surveying and Mapping. His research interests are in the areas of digital photogrammetry, remote sensing, image processing and pattern recognition.
Abstract:
Multi-temporal imagery change detection is growing in popularity among many applications, such as geography information updating, disaster monitoring, agriculture monitoring. With the improvement of spatial resolution, more subtle change information is expected to be detected. However, high-resolution imaging systems usually have low temporal resolution, resulting that multi-source images have to be considered to satisfy different kinds of applications, which brings increased challenges for change detection. Recently, deep learning is a fast-developing domain, making it possible for unsupervised abstract feature extraction of remote sensing images. For this reason, this paper proposed an unsupervised change detection approach using deep feature learning for high-resolution multi-temporal images acquired by different sensors. First, to obtain initial and reliable change information from bi-temporal images, multiple features are extracted including spectral features, texture features and edge features. Through utilizing these features jointly, specific rules are designed to select robust changed and unchanged samples automatically. Then, the bi-temporal multi-source images are layered as original change feature and Stacked De-noising Auto-Encoder (SDAE) is introduced for transforming the feature into a new feature space, where change information is represented deeply. Finally, the change detection model is constructed by adding a supervised classifier to the deeply learned features and the change information can be obtained by feeding the samples into the model with fine-tuning. Experiments with multi-temporal images from different sources demonstrate the effectiveness and robustness of the proposed approach.
Zhihui Gong
Zhengzhou University, China
Title: Joint salient feature and convolutional neural network for ship detection in remote sensing images
Time : 16:35-16:55
Biography:
Zhihui Gong is currently a full Professor at Department of Photogrammetry and Remote Sensing, Zhengzhou Institute of Surveying and Mapping. His main research interests include digital photogrammetry, remote sensing and image processing.
Abstract:
Due to the interference of clouds, sea waves, islands and other uncertain conditions on sea surface in satellite images, the majority of ship detection algorithms show poor performance in object detection and recognition. This paper proposes a method based on joint visual salient feature and convolutional neural network. First, the saliency map of image can be calculated by Phase spectrum of Fourier Transform (PFT), which is based on analysis of frequency domain. PFT can effectively suppress the interference of clouds and sea waves, but the distinction between background and ship is not notable. To solve this problem, adaptive logarithmic transformation is used to enhance the saliency map. Then, the gray morphological operation is adapted to eliminate noise areas and fill small holes. An adaptive image segmentation algorithm is used to extract all the salient area as the candidates. Finally, with a small number of ship samples and the idea of transferring learning to a simple convolutional neural network model can be trained. All candidate areas will be predicted by the model and the ships will be exactly detected and recognized. The experiments results show that our algorithm can effectively eliminate the interference of various factors such as cloud and islands and has the advantage on dealing with various kinds and scales of ships.
- Speaker Session: Robotics | Artificial Intelligence | Medical Robotics | Industrial Robot Automation | Humanoid Robots | Human-Robot Interaction | Deep Learning
Location: Polaris II
Chair
Emdad Khan
Internet Speech, USA
Session Introduction
Rattapon Thuangtong
Mahidol University, Thailand
Title: Robotic hair transplantation
Time : 13:45-14:15
Biography:
Rattapon Thuangtong is 2nd year PhD student in Biomedical Engineering, Faculty of Engineering, Mahidol University, Bangkok, Thailand and he is expertise in Hair transplantation-Robotic, FUT, FUE, Synthetic hair transplantation and he is professional doctor from Siriraj Hospital, Thailand
Abstract:
Follicular unit transplantation was classified into two techniques: (1) Strip harvesting follicular unit transplantation and (2) Follicular Unit Extraction (FUE). Artas robotic hair transplantation is the newly developed machine that use robotic arm to operate FUE. By using skin tensioner and photo sensor, robot machine can perceive the direction of each hair follicle within that frame and can process the robotic arm with two sets of punch (sharp punch and dull punch) perform FUE precisely. Artas robotic hair transplantation can improve FUE by increasing the speed and accuracy of harvesting. After using Artas robotic hair transplantation, for a few years, Artas function very well within the central area of the occiput that we called sweet spot. We found that there are some limitations of Artas robotic hair transplantation such as: (1) Limitation in the temporoparietal area of the scalp, (2) Limitation for the lower occipital area of the scalp (3) Limitation for female and senile patient, and (4) Artas has some limitation for previous surgery patients. We try to improve the result by: (1) Selecting the proper patient, not too old because of fragile tissue, (2) Perform tumescent solution injection as much as possible to decrease the would size and decrease the transection, (3) Upgrade the hardware and software of the Artas into 3,000 RPM version, (4) Combine using handheld motorized FUE machine to perform FUE in the temporoparietal and lower occipital, (5) Extract the hair follicles during on skin tensioner, and (6) Using team working to maximize the speed. Finally, robotic hair transplantation helps surgeon to perform FUE quicker and more accuracy. It is the important milestone on hair transplantation, but it needs more improvement for the best outcome of the patients.
Martin Heide Jørgensen
The Maersk Mc-Kinney Moller Institute-University of Southern Denmark, Denmark
Title: Industry 4.0, robotics and automation in the production environment: Future trends and challenges in product design
Time : 14:15-14:45
Biography:
Martin Heide Jørgensen is a Program Coordinator I4.0 since 2017 at University of Southern Denmark, The Maersk Mc-Kinney Moller Institute. He had completed his PhD from Aalborg University in the area of Fracture Mechanics. Later, he has contested different jobs, all in the area of public research and education, including a period of 7 years as Head of Department at Aalborg University. Within the last 3-4 years, his research area has been digitalization and I4.0.
Abstract:
Together with the digitalization, the frames for product design are changing substantial. Tools for simulation and digital twin representations now can be connected directly with the CAD systems. This means that the optimization of the product or the mechanical solution now can be handled in a multi objective system involving performance, reliability, production technologies and value chain analysis. A very important feature is the possibility of including systematic performance studies of the use of the product or mechanical solution. This can be done still by empirically means, but due to the digitalization, also using systematic sensor input. This enables the possibility of a more user-oriented perspective in the design, but also that the design can be defined in a more open manner, where the customer by digital tools can define some free elements and functionalities within a given frame of design freedom. To realize these new possibilities a more flexible and agile production system is required. In this sense the number of robot, flexible production units and new digitalized technologies are introduced in the production environments. This leads off cause to a higher degree of flexibility, but also a need for planning and control to obtain an economic feasible productivity. For many industrial CEO’s there is a big challenge in finding the right strategy and track in the world of digitalization to balance the cost and productivity with the ability to act agile and in harmony when new marked possibilities occur. The business strategy and strategy for optimizing of products and production setup is getting more complex and specific for the individual company. The challenge is to find some generic tools or methods to support this development.
Aleksei Yuzhakov
Promobot LLC, Russia
Title: Promobot: Autonomous service robot for business
Time : 14:45-15:15
Biography:
Alexey Yuzhakov has completed his PhD at the Pern National Research University. He is the CEO and Chairman of the board of directors of Promobot, a service robotics company.
Abstract:
Promobot is an autonomous service robot for your business. It is designed to work in crowded spaces, where, by moving autonomously, it helps people with navigation, communicates and answers any questions, shows promotional materials and remembers everyone with whom it interacted. Promobot attracts the maximum audience to the company’s products, as well as takes the people out of the process since it works autonomously. Today, several hundred of Promobot robots work in 26 countries, almost on every continent. They work as administrators, promoters, hosts, museum guides in such companies as NPF Sberbank, Beeline, Museum of Contemporary History of Russia, the Moscow Metro, and are able to increase financial performance of companies, quality of service and customer loyalty. Combination of the technologies of speech recognition, face recognition, speech synthesis, integrated linguistic database and ability to look up the information on the Internet allowed us to form a unique experience for the customers. When they come up to Promobot they see not only a voice assistant, ATM machine, toy or a terminal, they see a robot and they have to communicate with it like with a robot. And the robot has to communicate with them, answering their questions, helping them, showing them the way, giving them a pass, etc. Promobot is not a part of something fancy which you only see on Youtube, it is a part of our life and it is time to face it.
- Workshop
Location: Polaris II
Chair
Emdad Khan
Internet Speech, USA
Session Introduction
Caner Sahin
Imperial College London, UK
Title: Category-level 6D object pose recovery in depth images
Time : 15:15-16:00
Biography:
Caner Sahin is a PhD student in Imperial Computer Vision and Learning Lab at the Department of Electrical and Electronic Engineering of Imperial College, London. His PhD research is based on computer vision and machine learning. Particularly, he is working on object recognition, detection and 6D pose estimation.
Abstract:
Intra-class variations, distribution shifts among source and target domains are the major challenges of category level tasks. In this study, we address category level full 6D object pose estimation in the context of depth modality, introducing a novel part-based architecture that can tackle the above mentioned challenges. Our architecture particularly adapts the distribution shifts arising from shape discrepancies and naturally removes the variations of texture, illumination, pose, etc. so we call it as Intrinsic Structure Adaptor (ISA). We engineer ISA based on the innovations: (1) Semantically Selected Centers (SSC) are proposed in order to define the 6D pose at the level of categories, (2) 3D skeleton structures, which we derive as shape-invariant features are used to represent the parts extracted from the instances of given categories and privileged one-class learning is employed based on these parts, (3) graph matching is performed during training in such a way that the adaptation/generalization capability of the proposed architecture is improved across unseen instances. Experiments validate the promising performance of the proposed architecture.
Biography:
Suranjana Trivedy is currently a Faculty in GATE IIT Training Institute, India. She was Research Scholar in IIIT Hyderabad. Her current research interests are in the area of artificial intelligence, building intelligent system, robotics, autonomous vehicle and UAV.
Abstract:
The aim of the project is to build an intelligent naturally dialog-able machine, which can talk with the depressed and lonely older people and can finally become their true companion. Specifically, the research goal was to build an automatic system to understand multimodal emotion, including facial expressions, speech tone and the linguistic emotions. For that purpose we used semi-supervised learning methods, standard Natural Language Processing (NLP) (uni-gram) and Information Retrieval (IR) Term-Frequency Inverse-Document-Frequency (TF-IDF) techniques to analyze emotion from text. We have extracted emotion from facial images. Overall aim is to create intelligent companion bot.