본문 바로가기

전체 글

Framer: Interactive Frame Interpolation https://arxiv.org/abs/2410.18978 Framer: Interactive Frame InterpolationWe propose Framer for interactive frame interpolation, which targets producing smoothly transitioning frames between two images as per user creativity. Concretely, besides taking the start and end frames as inputs, our approach supports customizing the traarxiv.org 초록우리는 상호작용 프레임 보간을 위한 "Framer"를 제안합니다. 이는 사용자의 창의성에 따라 두 이미지.. 더보기
Magentic-One: A Generalist Multi-Agent System for Solving Complex Tasks https://www.microsoft.com/en-us/research/articles/magentic-one-a-generalist-multi-agent-system-for-solving-complex-tasks/ https://github.com/microsoft/autogen/tree/main/python/packages/autogen-magentic-one autogen/python/packages/autogen-magentic-one at main · microsoft/autogenA programming framework for agentic AI 🤖. Contribute to microsoft/autogen development by creating an account on GitHub... 더보기
AI 캘리포니아의 '인공지능' 공식 표준 정의 https://www.youtube.com/watch?v=8yLOocrrwvc 캘리포니아 주의 법률 AB 285에 따라, 인공지능은 다음과 같이 정의되었습니다:자율성의 정도가 다양한 엔지니어링 또는 기계 기반 시스템으로,명시적 또는 암시적 목표를 위해 입력받은 정보를 처리하여,물리적 또는 가상 환경에 영향을 미칠 수 있는 출력을 생성하는 시스템.1. 법안 AB 285목적: 급변하는 AI 기술 환경에서 캘리포니아 법률 간 일관성을 유지하고 명확한 AI 규제 체제를 수립.주요 내용:AI를 자율성과 정보처리 능력을 가진 시스템으로 정의.향후 모든 캘리포니아 AI 관련 입법에 공식 표준 정의를 적용.2. 캘리포니아 온라인 커뮤니티 칼리지 지침내용: AI, 데이터 과학, 머신러닝을 활용해 학생 지원책을 개발하도록.. 더보기
Scalable watermarking for identifying large language model outputs https://huggingface.co/blog/synthid-text Introducing SynthID TextIntroducing SynthID Text Do you find it difficult to tell if text was written by a human or generated by AI? Being able to identify AI-generated content is essential to promoting trust in information, and helping to address problems such as misattributionhuggingface.co 요약대규모 언어 모델(LLMs)은 사람의 글과 구분하기 어려울 정도로 고품질의 합성 텍스트를 대량으로 생성할 수 .. 더보기
X-Portrait 2: Highly Expressive Portrait Animation https://byteaigc.github.io/X-Portrait2/ X-Portrait 2: Highly Expressive Portrait AnimationPortrait animation technology provides a ultra-low cost and highly effective way to creating expressive, realistic character animations and video footages: users only need to provide a static portrait image and a driving performance video, and the model cabyteaigc.github.io 초상화 애니메이션 기술초상화 애니메이션 기술은 표현력 있고 .. 더보기
TANGO: Co-Speech Gesture Video Reenactment with Hierarchical Audio Motion Embedding and Diffusion Interpolation https://arxiv.org/abs/2410.04221 TANGO: Co-Speech Gesture Video Reenactment with Hierarchical Audio Motion Embedding and Diffusion InterpolationWe present TANGO, a framework for generating co-speech body-gesture videos. Given a few-minute, single-speaker reference video and target speech audio, TANGO produces high-fidelity videos with synchronized body gestures. TANGO builds on Gesture Video Ree.. 더보기
Simplifying, Stabilizing and Scaling Continuous-Time Consistency Models https://arxiv.org/abs/2410.11081 Simplifying, Stabilizing and Scaling Continuous-Time Consistency ModelsConsistency models (CMs) are a powerful class of diffusion-based generative models optimized for fast sampling. Most existing CMs are trained using discretized timesteps, which introduce additional hyperparameters and are prone to discretization errors. Wharxiv.org 요약일관성 모델(Consistency Models,.. 더보기
Pyramidal Flow Matching for Efficient Video Generative Modeling https://arxiv.org/abs/2410.05954 Pyramidal Flow Matching for Efficient Video Generative ModelingVideo generation requires modeling a vast spatiotemporal space, which demands significant computational resources and data usage. To reduce the complexity, the prevailing approaches employ a cascaded architecture to avoid direct training with full resolutiarxiv.org 초록비디오 생성은 광범위한 시공간 공간을 모델링해야 하며, 이는 .. 더보기