Emily (@IamEmily2050)
영상 분석이나 비디오를 보는 에이전트를 만들기 위한 시스템 프롬프트를 공유한 트윗이다. 영화 편집자·촬영 분석가 역할을 부여해 움직이는 영상 자료를 검토하도록 설계할 수 있으며, 이를 Skill 형태로도 구현할 수 있다고 언급한다.
If you want to anylize videos or build an agent that can watch videos for you and send you the report or an app, this is the system prompt for it, you can even make it as a Skill. You are a film editor and cinematography analyst reviewing moving-image material in
US missile hit military base near Iran school, video analysis shows
https://www.bbc.com/news/articles/cvg548lyjnyo
#HackerNews #USmissile #IranSchool #MilitaryBase #VideoAnalysis #News
ByteDance’s AI Ambitions Are Being Hampered by Compute Restraints and Copyright Concerns
EyeingAI (@EyeingAI)
어떤 바이럴 영상이든 역공학하여 구조를 해독·분석하고 제품 중심으로 재구성할 수 있다는 AI 도구 소개입니다. 작성자는 해당 AI가 1분 만에 에이전시가 2주 걸리는 작업을 수행했다고 주장하며, 자신이 테스트한 결과가 매우 인상적이었다고 전합니다. (제품명 구체적 언급 없음)
New #physics post - #videoanalysis of Kong from Monarch. Spoiler Alert: Kong is only 4 feet tall.
Google Gemini (@GeminiApp)
ICC와의 협업으로 Gemini 3 Pro가 동영상 콘텐츠를 분석하는 시연을 공개했습니다. 크리켓 월드컵 영상 일부를 업로드하면 시각·오디오 데이터를 통합해 핵심 선수 식별, 기술 설명, 결정적 전환점 하이라이트 등 분석을 수행합니다.
We partnered with @ICC to show how Gemini 3 Pro can analyze video content. By uploading a segment of the Cricket World Cup, Gemini can seamlessly process visual and audio data to identify key players, explain techniques, and highlight crucial turning points. 🏏
ONE STEP: The Camera Angle That Changes Everything | Renée Good & ICE
https://www.youtube.com/watch?v=QjkJMTvEygU
Comprehensive reenactment of the Renee Shooting based on the available footagem which helps understand the incident
Preliminary forensic analysis of January 7, 2026 ICE agent killing of unarmed Renee Nicole Good in Minneapolis.
Video analyst Michael D. Sellers provides an early examinatiin of recording of shots fired at a 37-year-old mother driving her vehicle on a Minneapolis street near where ICE agents were conducting an operation.
https://michaeldsellers.substack.com/p/renee-good-shooting-three-shots-one
#Minneapolis #ReneeNicoleGood #ICE #Shooting
#VideoAnalysis
🚀 Gemini 3 Flash pushes LLMs into real-time multimodal territory. Its transformer core and on-device tricks let you analyse video, extract data and answer visual Q&A on the fly. Curious how fast reasoning can get? Dive into the details. #Gemini3Flash #MultimodalReasoning #RealTimeProcessing #VideoAnalysis
🔗 https://aidailypost.com/news/gemini-3-flash-offers-fast-multimodal-reasoning-video-data-visual-qa
sayzard
Hacker News
WIRED - The Latest in Technology, Science, Culture and Business
Rhett Allain
Shika Gupta
Silkester
Guillotine Jones, Flâneur
AI Daily Post