登录
首页 » Python » mx-maskrcnn-master

mx-maskrcnn-master

于 2020-06-17 发布
0 251
下载积分: 1 下载次数: 5

代码说明:

说明:  我们提出了一个简单、灵活和通用的对象实例分割框架。我们的方法能有效检测图像中的对象,同时为每个实例生成高质量的 segmentation mask。这种被称为 Mask R-CNN 的方法通过添加用于预测 object mask 的分支来扩展 Faster R-CNN,该分支与用于边界框识别的现有分支并行。Mask R-CNN 训练简单,只需在以 5fps 运行的 Faster R-CNN 之上增加一个较小的 overhead。此外,Mask R-CNN 很容易推广到其他任务,例如它可以允许同一个框架中进行姿态估计。我们在 COCO 系列挑战的三个轨道任务中均取得了最佳成果,包括实例分割、边界对象检测和人关键点检测。没有任何 tricks,Mask R-CNN 的表现优于所有现有的单一模型取得的成绩,包括 COCO 2016 挑战赛的冠军。(Mask R-CNN code by HeKaiming)

文件列表:

mx-maskrcnn-master, 0 , 2018-02-28
mx-maskrcnn-master\.gitignore, 988 , 2018-02-28
mx-maskrcnn-master\.gitmodules, 103 , 2018-02-28
mx-maskrcnn-master\LICENSE, 11357 , 2018-02-28
mx-maskrcnn-master\Makefile, 221 , 2018-02-28
mx-maskrcnn-master\README.md, 5451 , 2018-02-28
mx-maskrcnn-master\data, 0 , 2018-02-28
mx-maskrcnn-master\data\cityscape, 0 , 2018-02-28
mx-maskrcnn-master\data\cityscape\imglists, 0 , 2018-02-28
mx-maskrcnn-master\data\cityscape\imglists\test.lst, 200205 , 2018-02-28
mx-maskrcnn-master\data\cityscape\imglists\train.lst, 412545 , 2018-02-28
mx-maskrcnn-master\data\cityscape\imglists\val.lst, 67790 , 2018-02-28
mx-maskrcnn-master\demo_mask.py, 2115 , 2018-02-28
mx-maskrcnn-master\eval_maskrcnn.py, 2113 , 2018-02-28
mx-maskrcnn-master\figures, 0 , 2018-02-28
mx-maskrcnn-master\figures\maskrcnn_result.png, 900697 , 2018-02-28
mx-maskrcnn-master\figures\test.jpg, 40967 , 2018-02-28
mx-maskrcnn-master\incubator-mxnet, 0 , 2018-02-28
mx-maskrcnn-master\rcnn, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\CXX_OP, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\CXX_OP\roi_align-inl.h, 8596 , 2018-02-28
mx-maskrcnn-master\rcnn\CXX_OP\roi_align.cc, 2824 , 2018-02-28
mx-maskrcnn-master\rcnn\CXX_OP\roi_align.cu, 12308 , 2018-02-28
mx-maskrcnn-master\rcnn\CXX_OP\roi_align_v1-inl.h, 15877 , 2018-02-28
mx-maskrcnn-master\rcnn\CXX_OP\roi_align_v1.cc, 3090 , 2018-02-28
mx-maskrcnn-master\rcnn\CXX_OP\roi_align_v1.cu, 446 , 2018-02-28
mx-maskrcnn-master\rcnn\PY_OP, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\PY_OP\__init__.py, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\PY_OP\fpn_roi_pooling.py, 4584 , 2018-02-28
mx-maskrcnn-master\rcnn\PY_OP\mask_output.py, 1971 , 2018-02-28
mx-maskrcnn-master\rcnn\PY_OP\mask_roi.py, 2240 , 2018-02-28
mx-maskrcnn-master\rcnn\PY_OP\proposal_fpn.py, 8149 , 2018-02-28
mx-maskrcnn-master\rcnn\__init__.py, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\config.py, 5104 , 2018-02-28
mx-maskrcnn-master\rcnn\core, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\core\__init__.py, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\core\callback.py, 1710 , 2018-02-28
mx-maskrcnn-master\rcnn\core\loader.py, 24515 , 2018-02-28
mx-maskrcnn-master\rcnn\core\metric.py, 9044 , 2018-02-28
mx-maskrcnn-master\rcnn\core\module.py, 8588 , 2018-02-28
mx-maskrcnn-master\rcnn\core\solver.py, 3136 , 2018-02-28
mx-maskrcnn-master\rcnn\core\tester.py, 13716 , 2018-02-28
mx-maskrcnn-master\rcnn\cython, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\cython\.gitignore, 15 , 2018-02-28
mx-maskrcnn-master\rcnn\cython\__init__.py, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\cython\anchors.pyx, 1185 , 2018-02-28
mx-maskrcnn-master\rcnn\cython\bbox.pyx, 1763 , 2018-02-28
mx-maskrcnn-master\rcnn\cython\cpu_nms.pyx, 2241 , 2018-02-28
mx-maskrcnn-master\rcnn\cython\gpu_nms.hpp, 146 , 2018-02-28
mx-maskrcnn-master\rcnn\cython\gpu_nms.pyx, 1110 , 2018-02-28
mx-maskrcnn-master\rcnn\cython\nms_kernel.cu, 5064 , 2018-02-28
mx-maskrcnn-master\rcnn\cython\setup.py, 5515 , 2018-02-28
mx-maskrcnn-master\rcnn\dataset, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\dataset\__init__.py, 53 , 2018-02-28
mx-maskrcnn-master\rcnn\dataset\cityscape.py, 12991 , 2018-02-28
mx-maskrcnn-master\rcnn\dataset\ds_utils.py, 442 , 2018-02-28
mx-maskrcnn-master\rcnn\dataset\imdb.py, 13205 , 2018-02-28
mx-maskrcnn-master\rcnn\io, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\io\__init__.py, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\io\image.py, 5850 , 2018-02-28
mx-maskrcnn-master\rcnn\io\rcnn.py, 19628 , 2018-02-28
mx-maskrcnn-master\rcnn\io\rpn.py, 10379 , 2018-02-28
mx-maskrcnn-master\rcnn\io\threaded_loader.py, 20199 , 2018-02-28
mx-maskrcnn-master\rcnn\processing, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\processing\__init__.py, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\processing\assign_levels.py, 1221 , 2018-02-28
mx-maskrcnn-master\rcnn\processing\bbox_regression.py, 9983 , 2018-02-28
mx-maskrcnn-master\rcnn\processing\bbox_transform.py, 5023 , 2018-02-28
mx-maskrcnn-master\rcnn\processing\generate_anchor.py, 2443 , 2018-02-28
mx-maskrcnn-master\rcnn\processing\nms.py, 1414 , 2018-02-28
mx-maskrcnn-master\rcnn\pycocotools, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\pycocotools\UPSTREAM_REV, 80 , 2018-02-28
mx-maskrcnn-master\rcnn\pycocotools\__init__.py, 21 , 2018-02-28
mx-maskrcnn-master\rcnn\pycocotools\_mask.pyx, 11430 , 2018-02-28
mx-maskrcnn-master\rcnn\pycocotools\coco.py, 18296 , 2018-02-28
mx-maskrcnn-master\rcnn\pycocotools\cocoeval.py, 23849 , 2018-02-28
mx-maskrcnn-master\rcnn\pycocotools\mask.py, 4570 , 2018-02-28
mx-maskrcnn-master\rcnn\pycocotools\maskApi.c, 8249 , 2018-02-28
mx-maskrcnn-master\rcnn\pycocotools\maskApi.h, 2176 , 2018-02-28
mx-maskrcnn-master\rcnn\pycocotools\setup.py, 579 , 2018-02-28
mx-maskrcnn-master\rcnn\symbol, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\symbol\__init__.py, 30 , 2018-02-28
mx-maskrcnn-master\rcnn\symbol\symbol_mask_fpn.py, 33269 , 2018-02-28
mx-maskrcnn-master\rcnn\tools, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\tools\__init__.py, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\tools\demo_maskrcnn.py, 4732 , 2018-02-28
mx-maskrcnn-master\rcnn\tools\demo_single_image.py, 6421 , 2018-02-28
mx-maskrcnn-master\rcnn\tools\test_maskrcnn.py, 4730 , 2018-02-28
mx-maskrcnn-master\rcnn\tools\test_rpn.py, 4318 , 2018-02-28
mx-maskrcnn-master\rcnn\tools\train_maskrcnn.py, 9777 , 2018-02-28
mx-maskrcnn-master\rcnn\tools\train_rpn.py, 9360 , 2018-02-28
mx-maskrcnn-master\rcnn\utils, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\utils\__init__.py, 0 , 2018-02-28
mx-maskrcnn-master\rcnn\utils\combine_model.py, 709 , 2018-02-28
mx-maskrcnn-master\rcnn\utils\load_data.py, 1718 , 2018-02-28
mx-maskrcnn-master\rcnn\utils\load_model.py, 1999 , 2018-02-28
mx-maskrcnn-master\rcnn\utils\save_model.py, 762 , 2018-02-28
mx-maskrcnn-master\scripts, 0 , 2018-02-28
mx-maskrcnn-master\scripts\demo.sh, 509 , 2018-02-28
mx-maskrcnn-master\scripts\demo_single_image.sh, 432 , 2018-02-28

下载说明:请别用迅雷下载,失败请重下,重下不扣分!

发表评论

0 个回复

  • SINQ_release
    用于立体图像质量评价的无参考图像质量评价程序,运行在matlab环境下。(A no-reference image quality evaluation program for the quality evaluation of stereoscopic images, running in the MATLAB environment .)
    2018-03-13 09:26:19下载
    积分:1
  • Waver
    用VB写的画出波纹的程序(Written by VB painted corrugated procedures)
    2020-06-29 15:00:01下载
    积分:1
  • Rail-Lane-Lines-master (1)
    说明:  铁轨线检测,可以完成对直轨和弯轨的检测,效果较好(Rail line detection, can complete the detection of straight rail and curved rail, the effect is good)
    2021-03-08 14:39:28下载
    积分:1
  • bp-classification
    使用bp(反向传播算法)实现分类问题,并观察分类过程中误差的变化。(Use bp (back-propagation algorithm) to achieve the classification and observe the changes in classification process in the error.)
    2009-12-27 09:32:56下载
    积分:1
  • Harr特征的几种在matlab环境下的实现
    Harr特征的几种在matlab环境下的实现(Implementation of several Harr features in Matlab environment)
    2017-09-11 15:57:53下载
    积分:1
  • 123
    基于Hough 变换的SAR 图像舰船尾迹检测方法。该文引进了K均值聚类方法,将这些点集归类,归类后的每一类的类心能很好的代表一条尾迹,实验结果表明方法是有效的。(Hough transform based on SAR images of ship wake detection. The introduction of the K-means clustering method to classify these points, is classified in each category after the class on behalf of the heart can be a good wake, experimental results show that the method is effective.)
    2009-05-14 15:50:37下载
    积分:1
  • 11_人脸检测与识别系统
    1.本程序主要处理BMP位图。 2.为了方便读者学习,程序各步骤都被拆分开实现了,通过点击相应按钮可以看到每个步骤处理的结果。 3.如果是进行人脸识别,需要先打开一幅位图依次进行“人脸检测与定位”操作和“人脸特征标注”操作,然后再打开需要比较的第二幅位图,也是依次进行所有操作,最后进行识别以区分先后两个人脸。(This procedure deals mainly with BMP Bitmaps. 2. In order for the reader to learn, each step of the program is split and the results of each step can be seen by clicking the appropriate button. 3. In the case of face recognition, it is necessary to open a bitmap for "face detection and location" and "face feature tagging" in turn, then open the second bitmap to be compared, and then perform all the operations in turn, and finally identify to distinguish two faces in succession.)
    2018-02-24 09:12:44下载
    积分:1
  • high-precision-laser-ranging-method
    测距应用广泛,普通标尺测距很难达到精微测距的要求,激光在这方面有着明显的优势,本文更是运用相位来进一步提高检测的精度!(Ranging widely, ranging ordinary scale ranging requirements difficult to achieve subtle, laser has a distinct advantage in this regard, this paper is the use of phase to further improve the detection accuracy!)
    2013-11-02 17:15:23下载
    积分:1
  • 道路裂缝识别算法
    本文提出了一种新的检测算法以适应不同路面,不同光照 的情况。本算法首先对图像进行预处理操作,随后使用神经网络 对预处理得到后的图像进行裂缝区域的识别,最终从识别结果图中提 取出裂缝信息(Research on road crack algorithm)
    2017-10-20 14:44:30下载
    积分:1
  • P-LDA
    P-LDA算法的代码,2009年的最新刊物的文章,较新的改进算法,供大家学习交流!(Perturbation LDA)
    2009-02-25 10:43:11下载
    积分:1
  • 696516资源总数
  • 106918会员总数
  • 4今日下载