# Interesting papers ## CoVLA: Comprehensive Vision-Language-Action Dataset for Autonomous Driving - https://arxiv.org/pdf/2408.10845v1 <img width="1615" alt="image" src="https://github.com/user-attachments/assets/dd1280e6-618b-46b9-af22-fe44707941da"> # Gotta go fast ## KFAST - Rosten 2006 의 FAST feature를 AVX2 가속 - 기존 방식에 비해 6 배 빠름 - https://github.com/komrad36/KFAST ## RGB2Y - rgb2gray - OpenCV 보다 2-4배 빠름 - https://github.com/komrad36/RGB2Y ## KfNN - float vector Nearest neighbor matching (SIFT 매칭할 때...) - https://github.com/komrad36/KfNN ## FeatureAngle - Fast feature angle direction 계산할 때 - https://github.com/komrad36/FeatureAngle ## KLERP - Bilinear interpolation을 OpenCV 보다 빠르게! - https://github.com/komrad36/KLERP ## K2NN - Brute force matching을 빠르게 - https://github.com/komrad36/K2NN ## BoxBlur - Box blurring을 빠르게! - https://github.com/komrad36/BoxBlur --- # 가짜연구소 - Gaussian Belief Propagation, Coded-SLAM에 관심 있으신 분들 - 텐스토렌트 서버에서 LLM 추론하는거 보고 싶으신 분들 - https://umoh.io/ko/pseudo-1st-gg <img width="750" alt="image" src="https://github.com/user-attachments/assets/cc3f7b48-49be-429f-90ca-b7a624dc099c">