https://drfirstlee.github.io/blog/Exploring AI Agents, LVLM, Multi-Modal, Robotics — a learning journey by DrFirst. 2026-07-08T04:44:22+09:00 Drfirst https://drfirstlee.github.io/blog/ Jekyll © 2026 Drfirst /blog/assets/img/favicons/favicon.ico /blog/assets/img/favicons/favicon-32x32.png 2026-06-12T09:00:00+09:00 2026-06-12T09:00:00+09:00 https://drfirstlee.github.io/blog/posts/SHOE/ DrFirst

👟 SHOE 논문 읽기! 논문: SHOE: Semantic HOI Open-Vocabulary Evaluation Metric 저자: Maja Noack, Qinqian Lei, Taipeng Tian, Bihan Dong, Robby T. Tan, Yixin Chen, John Young, Saijun Zhang, Bo Wang 학회: CVPR 2026 Workshop on Grounding and Reasoning with Vision-Language Models (GRAIL-V) 코드: https://github.com/majnoa/SHOE 한 줄 요약: Open-vocabulary HOI 예측에서 lean on couch와 sit on couch처럼 표현은 다르지만 의미적으로 가까운 답을...

2026-06-06T09:00:00+09:00 2026-06-06T09:00:00+09:00 https://drfirstlee.github.io/blog/posts/CrossHOI/ DrFirst

🔀 CrossHOI-Bench 논문 읽기! 논문: CrossHOI-Bench: A Unified Benchmark for HOI Evaluation across Vision-Language Models and HOI-Specific Methods 저자: Qinqian Lei, Bo Wang, Robby T. Tan 소속: National University of Singapore, University of Mississippi, ASUS Intelligent Cloud Services 학회: CVPR 2026 코드: https://github.com/ChelsieLei/CrossHOI-Bench 프로젝트 페이지: https://chelsielei.github.io/crosshoibench_pag...

2026-05-27T09:00:00+09:00 2026-05-27T09:00:00+09:00 https://drfirstlee.github.io/blog/posts/AVABench/ DrFirst

🧩 AVA-Bench — 핵심 논문 리포트 논문: AVA-Bench: Atomic Visual Ability Benchmark for Vision Foundation Models 저자: Zheda Mai, Arpita Chowdhury et al. (The Ohio State University, Adobe Research, Boston University) 학회: CVPR 2026 Project Page : https://zheda-mai.github.io/AVA-Bench/ 핵심 요약: 복잡한 VQA 점수만으로는 Vision Foundation Model(VFM)의 진짜 능력을 낱낱이 파헤칠 수 없다! 14가지 원자적 시각 능력(Atomic Visual Abilities)으로 완전히 분해하여...

2026-05-13T09:00:00+09:00 2026-05-13T09:00:00+09:00 https://drfirstlee.github.io/blog/posts/BARISTA/ DrFirst

☕ BARISTA — 핵심 논문 리포트 논문: BARISTA: A Multi-Task Egocentric Benchmark for Compositional Visual Understanding 저자: Patrick Knab, Orgest Xhelili et al. 학회/저널: arXiv 2026 (Submitted on 13 May 2026) Dataset/Code : Hugging Face - ramblr/BARISTA 핵심 요약: 인공지능이 커피를 직접 내릴 수 있을까? 단순히 커피를 보는 것을 넘어 사물 추적, 손과 물체의 상호작용, 행동 분석, 그리고 시간적 인과관계까지 한 번에 진단하는 최초의 고밀도 바리스타 비디오 벤치마크 BARISTA 등장! 🧩 문제 정의: 인공지능은 1인칭...

2026-05-08T09:00:00+09:00 2026-05-08T09:00:00+09:00 https://drfirstlee.github.io/blog/posts/GroupHOI/ DrFirst

👥 GroupHOI: Learning Human-Object Interaction as Groups 논문 읽기! 제목: Learning Human-Object Interaction as Groups 프로젝트: https://github.com/JiajunHong1/GroupHOI 저자: Jiajun Hong, Jianan Wei, Wenguan Wang 소속: Zhejiang University 발표: NeurIPS 2025 README citation 기준: The Thirty-ninth Annual Conference on Neural Information Processing Systems 핵심 키워드: Human-Object Interaction Detection, Gro...