-
Notifications
You must be signed in to change notification settings - Fork 0
/
content.json
1 lines (1 loc) · 6.76 KB
/
content.json
1
{"meta":{"title":"Xin Huang - HOME","subtitle":"What hurts more,the pain of hardwork or the pain of regret?","description":"An undergraduate student of SCUT","author":"Xin Huang","url":"http://yoursite.com"},"pages":[{"title":"","date":"2018-02-04T11:35:43.691Z","updated":"2018-02-04T09:55:36.141Z","comments":true,"path":"A Survey on Object Detection and Segmentation.html","permalink":"http://yoursite.com/A Survey on Object Detection and Segmentation.html","excerpt":"","text":"A Survey on Object Detection and Segmentation0.资料 后RCNN时代的物体检测及实例分割进展 RCNN 是近年来物体检测的基石,有大量的论文和博客在分析和改进 RCNN,本篇文章希望通过介绍除了 RCNN,YOLO,SSD 和 Mask RCNN 之外的 novel idea 来帮助大家开拓眼界,避免思维陷入单一模式。 2018-1 Speed/accuracy trade-offs for modern convolutional object detectors 2016-11 The goal of this paper is to serve as a guide for selecting a detection architecture that achieves the right speed/memory/accuracy balance for a given application and platform. 从近两年的CVPR会议来看,目标检测的研究方向是怎么样的? - Old Xie的回答 - 知乎 2017-3 目前object detection的工作可以粗略的分为两类:1:使用region proposal的,目前是主流,比如RCNN、SPP-Net、Fast-RCNN、Faster-RCNN以及MSRA最近的工作R-FCN2:不使用region proposal的,YOLO,SSD Panoptic Segmentation 2018-1 We propose and study a novel ‘Panoptic Segmentation’ (PS) task. Panoptic segmentation unifies the traditionally distinct tasks of instance segmentation (detect and segment each object instance) and semantic segmentation (assign a class label to each pixel). A 2017 Guide to Semantic Segmentation with Deep Learning 2017-7 In this post, I review the literature on semantic segmentation. Most research on semantic segmentation use natural/real world image datasets. Although the results are not directly applicable to medical images, I review these papers because research on the natural images is much more mature than that of medical images. 1. 工作目标-快速物体检测 speed/accuracy trade-off mobile设备只用来演示 旨意:在grid cell粗粒度的segmentation,在后期进行细粒度的或者直接参考yolov2直接输出结果? extention of YOLOv2 2.工作记录Week 1-2: 读各类paper,环境配置Week 3-4: 到实验室工作,重新配置环境,尝试将各网络交由mobile设备进行inferenceWeek 5-6: 3.代码解析 mask r-cnn YOLOv2 keras ?"},{"title":"","date":"2018-02-05T06:24:12.388Z","updated":"2018-02-04T14:23:54.654Z","comments":true,"path":"google8a44534199e20a34.html","permalink":"http://yoursite.com/google8a44534199e20a34.html","excerpt":"","text":"google-site-verification: google8a44534199e20a34.html"},{"title":"","date":"2018-02-06T05:10:32.080Z","updated":"2018-02-06T05:08:15.789Z","comments":true,"path":"recent work.html","permalink":"http://yoursite.com/recent work.html","excerpt":"","text":"Recent workSpeedy object detectionrelated work: yolov2 mask r-cnn instance segmentation 任务1.add a instance segmentation branch to yolov2 在yolo划分出来的每个小grid中进行一部分的seg(感觉不太科学),总之就是加seg,且要求speed/acc trade-off合理yolov2是single stage的方法,同SSD,而RCNN类的都是two stage 的方法,想要学习mask rcnn中的方法给yolo加个segmentation的branch,感觉有点难? todo-list: read and analyse the paper of YOLOv2 and mask rcnn read and analyse the code of YOLOv2 and mask rcnn which inplemented by Pytorch or TF(hard) try to add a instance segmentation branch to YOLOv2,like mask rcnn do to faster rcnn"},{"title":"Categories","date":"2018-02-04T12:30:53.000Z","updated":"2018-02-05T06:27:33.161Z","comments":true,"path":"categories/index.html","permalink":"http://yoursite.com/categories/index.html","excerpt":"","text":""},{"title":"Tags","date":"2018-02-04T12:30:32.000Z","updated":"2018-02-05T04:12:42.543Z","comments":true,"path":"tags/index.html","permalink":"http://yoursite.com/tags/index.html","excerpt":"","text":""}],"posts":[{"title":"Schedule of recent work","slug":"Recent-work","date":"2018-02-06T05:11:49.000Z","updated":"2018-02-07T10:40:33.715Z","comments":true,"path":"2018/02/06/Recent-work/","link":"","permalink":"http://yoursite.com/2018/02/06/Recent-work/","excerpt":"","text":"Recent workSpeedy object detectionrelated work: yolov2 mask r-cnn instance segmentation 任务1.add a instance segmentation branch to yolov2 在yolo划分出来的每个小grid中进行一部分的seg(感觉不太科学),总之就是加seg,且要求speed/acc trade-off合理yolov2是single stage的方法,同SSD,而RCNN类的都是two stage 的方法,想要学习mask rcnn中的方法给yolo加个segmentation的branch,感觉有点难? todo-list: read and analyse the paper of YOLOv2 and mask rcnn read and analyse the code of YOLOv2 and mask rcnn which inplemented by Pytorch or TF(hard) try to add a instance segmentation branch to YOLOv2,like mask rcnn do to faster rcnn","categories":[{"name":"Study","slug":"Study","permalink":"http://yoursite.com/categories/Study/"},{"name":"Schedule","slug":"Study/Schedule","permalink":"http://yoursite.com/categories/Study/Schedule/"}],"tags":[{"name":"Schedule","slug":"Schedule","permalink":"http://yoursite.com/tags/Schedule/"}]},{"title":"Notes of Related work(YOLO & Mask R-CNN)","slug":"Notes-of-YOLOv2","date":"2018-02-06T05:10:54.000Z","updated":"2018-02-07T08:17:55.973Z","comments":true,"path":"2018/02/06/Notes-of-YOLOv2/","link":"","permalink":"http://yoursite.com/2018/02/06/Notes-of-YOLOv2/","excerpt":"","text":"Paper Notes of YOLO and Mask R-CNNMaterialYOLO: YOLO 论文阅读YOLO YOLOv2: YOLO: Real-Time Object Detection - Joseph Redmon YOLOv2 详解 YOLOv2 论文笔记 code: YOLOv2 in PyTorch RCNN: Mask R-CNN code:Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow","categories":[{"name":"Study","slug":"Study","permalink":"http://yoursite.com/categories/Study/"},{"name":"Notes","slug":"Study/Notes","permalink":"http://yoursite.com/categories/Study/Notes/"}],"tags":[{"name":"Study","slug":"Study","permalink":"http://yoursite.com/tags/Study/"},{"name":"Paper","slug":"Paper","permalink":"http://yoursite.com/tags/Paper/"},{"name":"Detection","slug":"Detection","permalink":"http://yoursite.com/tags/Detection/"}]},{"title":"你好,世界!","slug":"你好,世界","date":"2018-02-03T13:34:00.000Z","updated":"2018-02-05T05:21:29.018Z","comments":true,"path":"2018/02/03/你好,世界/","link":"","permalink":"http://yoursite.com/2018/02/03/你好,世界/","excerpt":"","text":"这是我在HEXO上的第一个POST 只是测试一下","categories":[{"name":"Test","slug":"Test","permalink":"http://yoursite.com/categories/Test/"}],"tags":[{"name":"Test","slug":"Test","permalink":"http://yoursite.com/tags/Test/"}]}]}