Earticle

가상 현실은 몰입감과 존재감을 극대화시킬 수 있다는 장점이 있어 비디오 게임, 교육, 치료 목적을 가진 다양한 콘텐츠에 널리 활용되고 있다. 또한 GearVR, DayDream, Occulus Quest 2 등의 HMD 등이 개발됨에 따라 가상 현실은 널리 보급되었다. 그러나 이러한 장비들을 사용하면 사용자가 실제 물리적인 세계를 볼 수 없다는 단점이 있어 실제 물체와 충돌하거나 넘어지는 위험한 상황에 빠질 수 있다. 만약 사용자가 시스 루 HMD를 사용하면 실제 물리적인 세계를 볼 수 있으므로 물체를 피할 수 있고 이러한 위험을 피할 수 있 는데 이러한 시스루 HMD는 광학 기반 방식이나 비디오 기반 방식으로 분류된다. 이러한 광학 기반 HMD는 렌즈를 통과하는 물리적 물체들을 볼 수 있지만 비용 문제가 있으며 비디오 기반 HMD는 비용과 카메라의 원본 이미지에 다양한 가상 효과를 추가할 수 있다는 장점이 있다. 본 논문은 이러한 장점들을 활용하여 저 렴한 비용으로 게임에 활용될 수 있도록 개발된 시스루 HMD를 설명하였으며 해당 HMD를 활용하는 게임 을 설명하였다. 본 논문을 통하여 개발된 비디오 시스루 HMD를 통해 더 많은 사용자가 저렴한 비용으로 혼합 현실 기반 게임을 즐길 수 있을 것으로 예상된다.

As Virtual Reality has the advantage that maximizes immersion and presence, it has been widely used for various contents such as video games, education, and therapy. In addition, the developments of see-closed HMDs such as GearVR, DayDream and Occulus Quest 2 accelerate these widespread. However, it still has the disadvantage that its users cannot see the physical world. Thus, the users can fall into dangerous situations that collide with one of the physical objects or fall. When the players use the see-through HMDs, they can avoid these dangers because the physical world is visible, and they can avoid the object. The see-through HMDs can use an optical-based and a video-based method. The optical-based HMD has advantages in that the physical objects can be directly seen but has a cost problem. The video-based HMD has advantages in the cost aspect and various virtual effects can be added to the raw images from the camera. For example, the virtual ball can be drawn in the real environment. To utilize these advantages for the game at a low cost, we developed the video see-through HMD only with the general smartphone and the affordable cardboard. We also developed a simple game to test the developed HMD. As a result of the affordable video see-through HMD, more users can play the MR-based games.

최근 소프트웨어 유통망(ESD)를 통해 유통되는 게임이 늘어남에 따라 사용자들이 원하는 게임을 찾는데 어 려움을 겪고 있다. 이에 따라 사용자들이 원하는 게임을 찾기 쉽게 태그를 생성하는 모델의 필요성이 대두 되고 있다. 본 논문에서는 태그를 생성하는 모델을 BERT를 통해 설계하였다. 태그 100개 중 가장 적합한 태 그를 4개 추출하기 위해 입력된 문장에 대해 각 태그별로 이진분류를 수행하고 이진분류 당시의 Softmax 값이 가장 컸던 태그 4개를 선택했다. 또한, 모델의 정확도를 위해서 약 33억 개의 다국어 단어로 학습한 pre-trained Multilingual BERT 모델과 약 5천만 개의 한국어 단어로 학습한 KoBERT 모델을 가져와 한국어 데이터로 학습(finetuning) 시켜 사용하였다. 실험에서 BERT 모델은 KoBERT 모델보다 F- 점수에서 9.19 % 더 나은 성능을 보입니다. 이는 언어 학습 데이터 세트의 크기가 특정 언어인 한국어 특성보다 더 중요하다는 것을 나타낸다.

As the number of games increases in the software distribution network (ESD), it is difficult to find the game that a user wants. Therefore, the game can be recommended based on some game keyword tags for the user. In this paper, we propose a method to automatically generate the game keyword tags from the game description with the deep learning model, BERT. To generate the appropriate game keyword tags, the proposed method extracts the 100 representative game keyword tags from a game publishing platform Steam, and it performs the binary classification per tag. Finally, it selects 4 game keyword tags with the highest Softmax scores. Considering the accuracy improvement, a Korean game description set is used for finetuning and optimization, so that it updates both the BERT model pretrained with approximately 3.3 billion multilingual words, and the KoBERT model pretrained with approximately 50 million Korean words. Experiments show that the BERT model performs 9.19 % at F-score better than the KoBERT model. It describes that the size of the training data set is much more important than the characteristics of the specific language.

VR 수술 시스템의 도입은 현대 의학 교육에 큰 의미가 있다. VR 수술 시스템은 의대생과 젊은 의사의 학습 효율을 크게 향상시키고 반복 훈련 목표를 달성하며 전통 의학 교육의 단점을 보완할 수 있다. 사회적 디지 털화가 심화됨에 따라 의료 분야의 VR 수술 시스템 구축이 시급해지고 있다. 본 논문은 의료에 적용되는 기능성 게임의 관점에서 VR 수술의 기능성게임 디자인을 연구한다. 먼저 사용자 경험 이론과 사용자 심리 적 필요사항을 바탕으로 기능성 게임에 대한 사용자 경험과 기대치를 분석하였다. 그리고 사용자의 실제 필 요사항과 게임 레벨 기능 디자인을 분석하고, 게임 디자인의 핵심 포인트를 강조하여 최적화하였다. 둘째, 3D MAX를 결합한 UE4 모델 구성의 주의할 점을 분석하였다. 마지막으로 위 연구를 바탕으로 흉강경 하에 서 폐기포 VR 수술의 본격적인 게임을 설계하였다. 게임 디자인은 대상 사용자를 분석하여 내용을 결정한 후 이를 전문의에게 확인하여 이루어졌다. 첫째, 인간 가슴의 얇은 CT 이미지를 기반으로 3DS MAX를 이용 하여 가슴 모델, 수술 도구 및 수술 장면을 시뮬레이션 하였다. 둘째, UE4 게임 엔진을 사용하였다. 이 기술 은 3D 모델이 기능성 게임 시뮬레이션의 요구 사항을 충족하도록 만들고 UE4 장면을 디자인하며 C ++ 언 어를 사용하여 게임 기능을 개발하였다.

The introduction of virtual surgery system is of great significance to modern medical education. Virtual operation system can significantly improve the study efficiency of medical students and young doctors, achieve the goal of repeated training, and make up for the shortcomings of traditional medical education. As social digitalization intensifies, the construction of virtual surgery system in the medical field is urgent. This paper studies the serious game design of VR surgery from the point of view of serious game applied to medicine. Based on the theory of user experience and user psychological needs, we first analyzed the user experience and expectations of serious games. We then analyzed the user's actual needs and the game level function design, and optimize the key points of game design that should be emphasized. Secondly, the precautions of model constructed with UE4 combined with 3D MAX are analyzed. Finally, based on the above research, the serious game of lung bulla VR operation under thoracoscope is designed. The game design content was determined by analyzing the target user, which then was confirmed by a professional doctor. Firstly, based on the thin CT image of the human chest, 3D MAX technology is used to simulate the chest model, surgical tool and surgical scene. Second, UE4 game engine is used. The technology makes 3D model to meet the requirements of serious game simulation, then designs the imported UE4 scene, develops the game functions in the research area using C++ language, and then packages and releases the related resources.

본 연구에서는 게임 영상과 같은 생성된 영상으로부터 물체를 인식하는 심층 학습 기반 모델의 성능을 향상 시키는 방법을 제시한다. 특히, 실제 영상으로 훈련된 물체 인식 모델에 대해서 게임 영상으로 추가 훈련을 수 행함으로써 물체 인식 성능이 향상됨을 검증한다. 본 연구에서는 심층 학습 기반의 물체 인식 모델들 중에서 가장 널리 사용되는 YoloV2 모델을 이용한다. 이 모델에 대해서 8 종류의 다양한 게임에서 샘플링한 160장의 게임 영상을 적용해서 물체 인식 모델을 다시 훈련하고, IoU와 정확도를 측정해서 본 연구에서 주장하는 게임 영상을 이용한 훈련이 효과적임을 입증한다.

We present a framework that improves the performance of deep learning-based object detection model for generated images including game scenes. In particular, we aim to verify that the additional training using images sampled from game scenes can improve the performance of the object detection model, which was pre-trained using photographs. Among the various object detection schemes including Yolo V1, Yolo V2 and SSD, we employ YoloV2 model, which is one of the most widely used deep learning-based object detection model. YoloV2 model is pretrained using diverse photographs. This model is further trained through 160 game scene images sampled from eight different kinds of games. We select the games that range from realistic scenes and highly deformed scenes. We measure IoU (intersection over union) and accuracy using this model. The comparison between our re-trained model and the original model demonstrates the effectiveness of our strategy.

과거 2015년 이전 게임소송 분쟁들은 게임물을 보호하기 위하여 지식재산권의 보호를 모색하는데 주로 쟁 점이 있었다. 게임 내 이미지, 캐릭터 등과 같이 게임 속 창작물은 저작권법에 의한 보호를 받을 수도 있다. 게임 내 로고, 심볼 혹은 표장은 상표권에 의한 보호를 받을 수도 있다. 그러나 게임의 명칭, 게임 규칙, 아 이템 기능 등과 같은 그 외의 게임요소들은 단독으로는 저작권이나 상표권 보호를 받기가 어려웠다. 최근, 게임 지적재산권 및 게임 사업 운영 관련 소송 과정에서 부정경쟁방지법 위반의 쟁점이 함께 제기되는 경 우가 발생하고 있다. 본고는 가장 최근인 2019년부터 2021년 5월까지 선고된 한국, 미국, 일본 게임 판례 중 부정경쟁방지법 쟁점을 다룬 사례들을 살펴본다. 최근 사례에 등장한 부정경쟁행위 유형은 크게 타인의 성과물을 무단 사용하고 경제적 이익을 해하는 일, 타인의 상품 혹은 그와 유사한 것의 표시를 자신의 영업 에 사용하는 것이다. 관련 분쟁이 독점거래 규제법 위반으로 확장되는 경우도 있다. 부정경쟁방지법에 의한 보호는 전통적인 지식재산법에 의하여 보호하기 어려운 영역도 포섭이 가능하다. 그러나 보호범위가 확장될 수록, 이는 한편으로 게임 개발자들이 새로운 게임을 제작할 경우에 타 게임의 창작물 혹은 게임 회사의 성 과물을 침해할 수 있는 위험이 증가함을 의미하기 때문에 주의할 필요가 있다.

Previously, the parties of the old video game litigations that alleged before 2015 in Korea seem to focus on seeking intellectual property rights to protect the works of games. Some creations in the game (i.e. images, game characters, etc.) might be protected by copyright law. The logo, symbol, or trademark of the game would be entitled to the Trademark rights. But the other features of the game, such as the title, game rule, and item functions, etc., were hard to be protected under neither laws above. Recently, violation of unfair competition is alleged with the other infringement claims in the lawsuits related to the game IP and businesses. This paper explores the very recent video game cases published between 2019 to May 2021 from the Courts of South Korea, the USA, and Japan. The types of violation of unfair competition prevention laws in such cases could be categorized as the defendant's activity of using achievements of other persons without their permission and harming their economic benefit or using the indication of other goods of it for one's own business. And sometimes, the case could become more serious turning into antitrust litigation. Protection under Unfair Competition Prevention Act could variously cover the issues that traditional IP laws cannot cover. But as the protection gets broader, the concern of game developers not to infringe other game's creation or game company's achievements also gets bigger when establishing a new game or a business.

In this study, we investigated whether a tool such as a game toy can be used as an augmented reality tool, and a system model that can be extended to a game element using wireless communication technology such as Bluetooth and a controllable module. This is an online ship type game using augmented reality technology and wireless communication technology. In addition, the existing game element was extended by applying a smartphone app control module. The existing game method uses the method of playing the game with only limited functions in the same space. This study expands to augmented reality-based games by implementing contents in a way that matches game objects with the grafting of augmented reality technology, and uses various items that emerge as the limit of reality. Therefore, we standardized the size of game objects so that they can be used three-dimension in all spaces on the screen according to the space arrangement such as overlapping prevention, distance, and height, and augmented reality technology was used to allow the game to be played by manipulation of a smartphone. In addition, we propose a system framework-based model that can be applied to various games, and a framework that can implement various augmented reality environments. The augmented reality-based battle game proposed in this study combines a knowledge-based augmented reality system that can be extended to game elements by modularizing the function of a toy through+ a context-aware agent based on context information and an intelligent DB based on domain knowledge.

Physical activity and exercise is an important step in the journey of a healthy lifestyle. It reduces the risk of developing many diseases like diabetes, cardiovascular diseases, and even cancer. In today’s world, we spend most of our time sitting in front of a computer or mobile phones for work and entertainment. This is one of the leading factors in a lack of physical activities and thus a decrease in the fitness level of many individuals. In this paper, we design a mobile 2D platformer game where progress inside the game world is dependent on physical activity in the real world. The game has two modes namely adventure and story mode. This design can motivate people to lead an active and healthy lifestyle. The physical activity monitoring is done solely through the pedometer of a smartphone held by the user while playing the game. The number of steps walked by the user determines the number of revives inside the adventure mode of the game. There are three levels in the story mode which are unlocked by walking more than a certain number of steps in the real world.

Mental health problems leading to depression have become a critical concern due to the growing engagement of people on social media platforms. Several past approaches have been implemented by analyzing the pattern and behaviour of the posts by users on social networking sites. This research study proposed a system for predicting users who may be depressed, based on the characteristics of users who is already affected. A combination of both the tweet-level and the user-level architecture was used to generate a more robust and reliable system where semantic embeddings trained from advanced neural networks were adopted under the tweet-level. SVM with Word2Vec and TF-IDF has been used and yielded an accuracy of 98.14% and recall of 95.63%.

In the following years, technology has progressed in so many ways that it has provided the cyber society with a resource that only computers can excel at, such as the art of counterfeit of media, which was before unavailable. Deepfakes are a term used to describe this kind of deception. The majority of well-documented Deep Fakes are produced using Generative Adversarial Network (GAN) Models, which are essentially two distinct Machine Learning Models that perform the roles of attack and defence. These models create and identify deepfakes until they reach a point where the morphing no longer detects the deepfakes anymore. Using this algorithm/model, it is possible to discover and create new media that has a similar demographic to the training set, resulting in the development of the ideal Deep Fake media. Because the alterations are carried out utilising advanced characteristics, they cannot be seen with the human eye. However, it is completely feasible to develop an algorithm that can automatically identify this kind of tampering carried out via the internet. This not only enables us to broaden the scope of our search beyond a single media item, but also beyond a large library of mixed media. The more it learns, the better it becomes as artificial intelligence takes over in full force with automation. In order to create better deep fakes, new models are being developed all the time, making it more difficult to distinguish between genuine and morphing material.