Notice
Recent Posts
Recent Comments
Link
일 | 월 | 화 | 수 | 목 | 금 | 토 |
---|---|---|---|---|---|---|
1 | ||||||
2 | 3 | 4 | 5 | 6 | 7 | 8 |
9 | 10 | 11 | 12 | 13 | 14 | 15 |
16 | 17 | 18 | 19 | 20 | 21 | 22 |
23 | 24 | 25 | 26 | 27 | 28 |
Tags
- 논문 구현
- 자료구조
- Python
- 큐
- NLP
- reconstruction
- Object Tracking
- Knowledge Distillation
- re-identification
- 임계처리
- Computer Vision
- deep learning
- level2
- center loss
- 스택
- 딥러닝
- 파이썬
- Threshold
- point cloud
- transformer
- OpenCV
- 3D
- flame
- Object Detection
- attention
- cv2
- 프로그래머스
- 알고리즘
- numpy
- Deeplearning
Archives
- Today
- Total
목록ICP (1)
공돌이 공룡의 서재
![](http://i1.daumcdn.net/thumb/C150x150/?fname=https://blog.kakaocdn.net/dn/YTGBA/btrdeuv1XOK/Rw1liELGsqvKI15rCRYSp0/img.png)
ICP: Iterative Closest Point 이 알고리즘은 두 개의 point cloud가 주어졌을 때, corresponding point 사이의 거리를 계산하여, 이 거리를 최소화하는 Transformation matrix을 찾는 알고리즘이다. 이 알고리즘을 공부하게 된 계기는 depth camera에서 reconstructed 된 point 들을 통해 camera간 pose를 계산하기 위함이었다. corresponding point를 찾는 알고리즘은 KDTree 등 여러 알고리즘이 있지만, 이건 다음에 살펴보고, 여기선 corresponding point들이 주어졌다는 가정하에 ICP 알고리즘에 대해서만 알아보겠다. 크게 point-to-point 방법과 point-to-plane 방법이 있..
컴퓨터비전
2021. 8. 26. 18:48