当前位置:网站首页>CVPR 2022 | querydet: use cascaded sparse query to accelerate small target detection under high resolution
CVPR 2022 | querydet: use cascaded sparse query to accelerate small target detection under high resolution
2022-04-23 20:11:00 【Zhiyuan community】
Although in the past few years , Universal target detection based on deep learning has achieved great success , However, the performance and efficiency of detecting small targets are far from satisfactory . The most common and effective method to promote small target detection is to use high-resolution images or feature maps . However , Both methods can lead to expensive calculations , Because the computational cost will increase with the increase of image and feature size .
We proposed QueryDet, it A novel query mechanism is used to speed up the inference speed of target detector based on feature pyramid . The pipeline It consists of two steps : Firstly, the coarse location of small targets is predicted on the low resolution features , Then the accurate detection results are calculated by using the high-resolution features of these coarse position sparse guidance . In this way, high resolution can be obtained feature map Of benefit, It can also avoid using less computation for the background area .
stay popular COCO On dataset , This method will mAP Improved 1.0,mAP-small Improved 2.0, The reasoning speed of high resolution is increased to 3.0×. In a that contains more small objects VisDrone On dataset , We have acquired new SOTA, At the same time, the average 2.3× High resolution acceleration .
Paper title :QueryDet: Cascaded Sparse Query for Accelerating High-Resolution for Small Object Detection
Thesis link :https://arxiv.org/abs/2103.09136
Code link :https://github.com/ChenhongyiYang/QueryDet-PyTorch
版权声明
本文为[Zhiyuan community]所创,转载请带上原文链接,感谢
https://yzsam.com/2022/04/202204232004207702.html
边栏推荐
- 指针数组与数组指针的区分
- [report] Microsoft: application of deep learning methods in speech enhancement
- 对普通bean进行Autowired字段注入
- DTMF双音多频信号仿真演示系统
- 【webrtc】Add x264 encoder for CEF/Chromium
- Deep learning -- Summary of Feature Engineering
- Mysql database - single table query (I)
- 如何在BNB链上创建BEP-20通证
- filebeat、logstash配置安装
- NC basic usage 1
猜你喜欢
Project training of Software College of Shandong University - Innovation Training - network security shooting range experimental platform (VII)
STM32 Basics
山东大学软件学院项目实训-创新实训-网络安全靶场实验平台(七)
如何在BNB鏈上創建BEP-20通證
WordPress插件:WP-China-Yes解决国内访问官网慢的方法
考研英语唐叔的语法课笔记
selenium.common.exceptions.WebDriverException: Message: ‘chromedriver‘ executable needs to be in PAT
@MapperScan与@Mapper
【文本分类案例】(4) RNN、LSTM 电影评价倾向分类,附TensorFlow完整代码
网络通信基础(局域网、广域网、IP地址、端口号、协议、封装、分用)
随机推荐
Openharmony open source developer growth plan, looking for new open source forces that change the world!
R语言使用timeROC包计算存在竞争风险情况下的生存资料多时间AUC值、使用cox模型、并添加协变量、R语言使用timeROC包的plotAUCcurve函数可视化多时间生存资料的AUC曲线
[2022] regard 3D target detection as sequence prediction - point2seq: detecting 3D objects as sequences
Possible root causes include a too low setting for -Xss and illegal cyclic inheritance dependencies
antd dropdown + modal + textarea导致的textarea光标不可被键盘控制问题
SRS deployment
filebeat、logstash配置安装
R language uses timeroc package to calculate the multi time AUC value of survival data under competitive risk, uses Cox model and adds covariates, and R language uses the plotauccurve function of time
程序设计语言基础(2)
IIS数据转换问题16bit转24bit
Physical meaning of FFT: 1024 point FFT is 1024 real numbers. The actual input to FFT is 1024 complex numbers (imaginary part is 0), and the output is also 1024 complex numbers. The effective data is
aqs的学习
Kubernetes entry to mastery - bare metal loadbalance 80 443 port exposure precautions
nc基础用法
网络通信基础(局域网、广域网、IP地址、端口号、协议、封装、分用)
Compact CUDA tutorial - CUDA driver API
MFC obtains local IP (used more in network communication)
Shanda Wangan shooting range experimental platform project - personal record (IV)
LeetCode动态规划训练营(1~5天)
指针数组与数组指针的区分