• Electronics and Telecommunications Trends
• /
• v.38 no.5
• /
• pp.1-11
• /
• 2023

The emergence of generative hyperscale artificial intelligence (AI) has enabled new services, such as image-generating AI and conversational AI based on large language models. Such services likely lead to the influx of numerous users, who cannot be handled using conventional AI models. Furthermore, the exponential increase in training data, computations, and high user demand of AI models has led to intensive hardware resource consumption, highlighting the need to develop domain-specific semiconductors for hyperscale AI. In this technical report, we describe development trends in technologies for hyperscale AI processors pursued by domestic and foreign semiconductor companies, such as NVIDIA, Graphcore, Tesla, Google, Meta, SAPEON, FuriosaAI, and Rebellions.

• International Journal of Internet, Broadcasting and Communication
• /
• v.15 no.2
• /
• pp.31-38
• /
• 2023

This paper introduces a real-time face swapping system that enables video influencers to swap their faces with arbitrary generated face images of their choice. The system is implemented as a Django-based server that uses a REST request to communicate with the generative model,specifically the pretrained stable diffusion model. Once generated, the generated image is displayed on the front page so that the influencer can decide whether to use the generated face or not, by clicking on the accept button on the front page. If they choose to use it, both their face and the generated face are sent to the landmark extraction module to extract the landmarks, which are then used to swap the faces. To minimize the fluctuation of landmarks over time that can cause instability or jitter in the output, a temporal filtering step is added. Furthermore, to increase the processing speed the system works on a reduced set of the extracted landmarks.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.443-445
• /
• 2022

인공지능 기술이 발전함에 따라 경찰의 범죄수사 분야에서도 인공지능 기술을 적용하고자 하는 연구가 활발하다. 범죄수사의 결과물인 수사결과 보고서 작성에 있어 판결문은 중요한 데이터가 될 수 있다. 그러나 판결문은 공개된 데이터의 이미지화로 인해 정형화된 데이터의 확보가 까다롭고, 소수의 법조계 전문가가 아닌 일반인이 생성해내기 어려워 데이터 확보가 쉽지 않은 현실이다. 이에 본 연구에서는 생성적 사전학습 언어모델을 이용한 판결문 문장 데이터 생성을 제안하였다. 카카오의 KoGPT를 활용하여 실제 판결문장 일부를 제시한 결과 판결문과 유사한 형태의 문장을 생성한 것을 확인하였다. 향후 판결문 데이터를 활용하기 위한 인공지능 기술 기반 범죄수사 연구에 있어, 생성된 판결문 데이터를 활용할 수 있을 것으로 기대된다.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.137-140
• /
• 2023

자연어처리는 인공지능 발전과 함께 주목받는 분야로 컴퓨터가 인간의 언어를 이해하게 하는 기술이다. 그러나 많은 인공지능 모델은 블랙박스처럼 동작하여 그 원리를 해석하거나 이해하기 힘들다는 문제점이 있다. 이 문제를 해결하기 위해 설명 가능한 인공지능의 중요성이 강조되고 있으며, 활발히 연구되고 있다. 연구 초기에는 모델의 예측에 큰 영향을 끼치는 단어나 절을 근거로 추출했지만 문제 해결을 위한 단서 수준에 그쳤으며, 이후 문장 단위의 근거로 확장된 연구가 수행되었다. 하지만 문서 내에 서로 떨어져 있는 근거 문장 사이에 누락된 문맥 정보로 인하여 이해에 어려움을 줄 수 있다. 따라서 본 논문에서는 사람에게 보다 이해하기 쉬운 근거를 제공하기 위한 생성형 기반의 근거 추론 연구를 수행하고자 한다. 높은 수준의 자연어 이해 능력이 필요한 문서 요약 데이터셋을 활용하여 근거를 생성하고자 하며, 실험을 통해 일부 기계독해 데이터 샘플에서 예측에 대한 적절한 근거를 제공하는 것을 확인했다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.297-298
• /
• 2023

COVID-19 가 사실상 종식됨에 따라 대학 강의가 비대면 온라인 강의에서 대면 강의로 전환되었다. 온라인 강의에서는 다시 보기를 통한 복습이 가능했지만, 대면강의에서는 녹음을 통해서 이를 대체하고 있다. 하지만 다시 보기와 녹음본은 원하는 부분을 찾거나 내용을 요약하는데 있어서 시간이 오래 걸리고 불편하다. 본 논문에서는 강의 내용을 STT(Speech-to-Text) 기술을 활용하여 텍스트로 변환하고 ChatGPT(Chat-Generative Pre-trained Transformer)로 요약하는 애플리케이션을 제안한다.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.2
• /
• pp.447-452
• /
• 2024

As the remarkable achievements of generative artificial intelligence technology become increasingly visible, the issue of 'controllability' in artificial intelligence is emerging as a prominent global keyword. This comes at a time when existential threats, such as the possibility of machines dominating humans, are being raised. Accordingly, this study aims to establish the groundwork for shaping a social public sphere by closely examining the concept of control, the current status, and the global landscape of artificial intelligence. It seeks to address the innovative changes anticipated in future society, with artificial intelligence technology at its core. The study aims to derive implications for preparing countermeasures against social problems and unpredictable variables that may arise from the evolution of artificial intelligence technology. It also aims to present guidelines and strategic insights for the establishment of government regulations. Furthermore, the study seeks to uncover implications for the formation of social public discourse.

• International Journal of Advanced Culture Technology
• /
• v.11 no.3
• /
• pp.225-230
• /
• 2023

With the development of AI technology, there is a growing discussion about Text-to-Image Generative AI. We presented a Generative AI video production method and delineated a methodology for the production of personalized AI-generated videos with the objective of broadening the landscape of the video domain. And we meticulously examined the procedural steps involved in AI-driven video production and directly implemented a video creation approach utilizing the VQGAN-CLIP model. The outcomes produced by the VQGAN-CLIP model exhibited a relatively moderate resolution and frame rate, and predominantly manifested as abstract images. Such characteristics indicated potential applicability in OTT-based video content or the realm of visual arts. It is anticipated that AI-driven video production techniques will see heightened utilization in forthcoming endeavors.

• Journal of the Korea Computer Graphics Society
• /
• v.28 no.2
• /
• pp.11-19
• /
• 2022

Recently, various researches on medical image generation have been suggested, and it becomes crucial to accurately evaluate the quality and diversity of the generated medical images. For this purpose, the expert's visual turing test, feature distribution visualization, and quantitative evaluation through IS and FID are evaluated. However, there are few methods for quantitatively evaluating medical images in terms of fidelity and diversity. In this paper, images are generated by learning a chest CT dataset of non-small cell lung cancer patients through DCGAN and PGGAN generative models, and the performance of the two generative models are evaluated in terms of fidelity and diversity. The performance is quantitatively evaluated through IS and FID, which are one-dimensional score-based evaluation methods, and Precision and Recall, Improved Precision and Recall, which are two-dimensional score-based evaluation methods, and the characteristics and limitations of each evaluation method are also analyzed in medical imaging.

• Journal of Digestive Cancer Research
• /
• v.12 no.1
• /
• pp.38-43
• /
• 2024

Artificial Intelligence (AI) is the intelligence of machines or software, in contrast to human intelligence. Generative AI technologies, such as ChatGPT, have emerged as valuable research tools that facilitate brainstorming ideas for research, analyzing data, and writing papers. However, their application has raised concerns regarding authorship, copyright, and ethical considerations. Many organizations of medical journal editors, including the International Committee of Medical Journal Editors and the World Association of Medical Editors, do not recognize AI technology as an author. Instead, they recommend that researchers explicitly acknowledge the use of AI tools in their research methods or acknowledgments. Similarly, international journals do not recognize AI tools as authors and insist that human authors should be accountable for the research findings. Therefore, when integrating AI-generated content into papers, it should be disclosed under the responsibility of human authors, and the details of the AI tools employed should be specified to ensure transparency and reliability.

• Korean Journal of Radiology
• /
• v.23 no.5
• /
• pp.500-504
• /
• 2022