質問するログイン新規登録

回答編集履歴

1

成形

2018/04/25 03:38

投稿

LouiS0616
LouiS0616

スコア35678

answer CHANGED
@@ -5,11 +5,7 @@
5
5
  [Fast Explicit Diffusion for Accelerated
6
6
  Features in Nonlinear Scale Spaces](http://www.bmva.org/bmvc/2013/Papers/paper0013/paper0013.pdf)
7
7
 
8
- > To preserve low computational demand and storage requirement, we also introduce a
9
- highly efficient Modified-Local Difference Binary (M-LDB) descriptor. While the original
10
- LDB descriptor presented in [20] is neither rotation nor scale invariant as BRIEF is, we build
11
- a robust binary descriptor that is rotation and scale invariant and exploits gradient information
12
- from the nonlinear scale space, increasing distinctiveness.
8
+ > To preserve low computational demand and storage requirement, we also introduce a highly efficient Modified-Local Difference Binary (M - LDB) descriptor. While the original LDB descriptor presented in [20] is neither rotation nor scale invariant as BRIEF is, we build a robust binary descriptor that is rotation and scale invariant and exploits gradient information from the nonlinear scale space, increasing distinctiveness.
13
9
 
14
10
  M-LDBは、LDBを回転不変およびスケール不変にしたものだそうです。
15
11