# Interesting papers ## MAGiC-SLAM - Multi-agent + Gaussian splatting SLAM - Submap merge -> Global map 전략 - https://arxiv.org/abs/2411.16785  ## Droid-Splat - Droid-SLAM + 3D Gaussian Splatting - Droid-SLAM? - Deep bundle adjustment + recurrent iterative pose & depth update - https://arxiv.org/abs/2411.17660  <img width="1636" alt="image" src="https://github.com/user-attachments/assets/b8c8821c-3d34-4356-878a-2107d8adc362"> ## Splat-AD - 이미지 + 라이다 렌더링 for 자율주행 - https://arxiv.org/abs/2411.16816  ## OVO-SLAM - Open-Vocabulary Online 3D semantic SLAM - Detect and track 3D segments + CLIP - https://arxiv.org/pdf/2411.15043 <img width="797" alt="image" src="https://github.com/user-attachments/assets/2701b6d1-3d16-4bf7-b59b-e70ef0f2da90"> <img width="1527" alt="image" src="https://github.com/user-attachments/assets/b8f8f48c-0df4-4408-bb27-42f06d03d6d1"> ## YOLO-cpp - https://github.com/Geekgineer/YOLOs-CPP - C++에서 바로 YOLO v5,7,8,10,11 사용 가능 - OpenCV와 함께 사용하기 쉬움  <img width="1253" alt="image" src="https://github.com/user-attachments/assets/5e9348b3-f8fa-4770-9664-c4f753409956"> ## ALIKED-cpp - https://github.com/MrNeRF/ALIKED_CPP - High-performance feature matching & tracking - Matches original descriptor performance - CUDA-accelerated inference - 50% faster - Simple API with OpenCV integration 