BING++: A Fast High Quality Object Proposal Generator at 100fps

最新推荐文章于 2024-06-04 09:43:28 发布

cv_family_z

最新推荐文章于 2024-06-04 09:43:28 发布

阅读量3.6k

点赞数

CC 4.0 BY-SA版权

分类专栏：目标检测 ZJ

本文链接：https://blog.csdn.net/cv_family_z/article/details/50259773

ZJ 同时被 2 个专栏收录

78 篇文章

订阅专栏

目标检测

43 篇文章

订阅专栏

针对BING算法在目标检测中存在候选区域框定位质量不佳的问题，本文提出BING++算法。该算法通过递归更新边界框并利用边缘特征来提高定位精度，并采用快速超像素合并技术进一步细化边界框。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

本文是对 BING 算法的升级，主要是在快的同时保持定位精度
两个 + 分别对应： edge-based recursive boxes as one “+”, and MTSE-based superpixel merging as the other “+”

Object Detection Recall (DR)：which is the ratio of the number of correctly detected objects and the total number of objects in the dataset

Proposal Localization Quality measured in terms of average best overlap (ABO)

mean average best overlap (MABO)
这里写图片描述

Key insight: 我们注意到好的候选区域框是刚刚包含物体，edge 特征可以用于检测物体边缘

这里写图片描述

2 Problems in BING
poor proposal localization quality

3 Our Solution: BING++
(1) We take proposals from BING as input.
(2) We then recursively update current bounding boxes based on their current locations and surrounding edge points to form new bounding boxes. This process is recursed until we find
(3) We finally apply fast super-pixel merging techniques to further refine the output bounding boxes of (2) and output final object proposals

算法流程：
这里写图片描述