High Performance Visual Tracking with Siamese Region Proposal Network 阅读笔记

1，(IDEA) In tracking task we don’t have pre-defined categories, so we need the template branch to encode the target’s appearance information into the RPN feature map to discriminate foreground from background.

2，(RPN) RPN has many successful applications in detection because of its speed and great performance, however, it hasn’t been fully exploited in tracking.

3，(NETWORK) we use the modified AlexNet where the groups from conv2 and conv4 are removed.

4，(KERNEL) The template feature maps [Math Processing Error][φ(z)]cls\left [ \varphi\left ( z \right ) \right ]_{cls} and [Math Processing Error][φ(z)]reg\left [ \varphi\left ( z \right ) \right ]_{reg} are used as kernels.

5，(LOSS) Softmax loss is adopted to supervise the classification branch. Loss for classification is the cross-entropy loss and we adopt smooth L1 loss with normalized coordinates for regression.

6，(DATA) During the training phase, sample pairs are picked from ILSVRC with a random interval and from Youtube-BB continuously. We extract image pairs from VID and Youtube-BB by choosing frames with interval less than 100 and performing further crop procedure

7，(TRAIN) We train Siamese-RPN end-to-end using Stochastic Gradient Descent (SGD) after the Siamese subnetwork being pretrained using Imagenet.

8，(AUGMENTATIONS) Because of the need of training regression branch, some data augmentations are adopted including affine transformation.

9，(SAMPLE) The criterion used in object detection task is adopted here that we use IoU together with two thresholds 0.6 and 0.3.

10，(SAMPLE) We also limit at most 16 positive samples and totally 64 samples from one training pair.

11，(TRICK) The first proposal selection strategy is discarding the bounding boxes generated by the anchors too far away from the center. We only keep the center 7×7 anchors.

12，(TRICK) The second proposal selection strategy is that we use cosine window and scale change penalty to re-rank the proposals’ score to get the best one.

13，(TRICK) After the final bounding box is selected, target size is updated by linear interpolation to keep the shape changing smoothly.

14，(TRAIN) We use a modified AlexNet pretrained from ImageNet with the parameters of the first three convolution layers fixed and only fine-tune the last two convolution layers in Siamese-RPN.

15，(TRAIN) There are totally 50 epoches performed and the learning rate is decreased in log space from 10−2 to 10−6.

16，(PLATFORM) Our experiments are implemented using PyTorch.

17，(ACCURACY) VOT2016 EAO：0.3441，OTB2015 AUC：0.637.

18，(SPEED) 160fps.

High Performance Visual Tracking with Siamese Region Proposal Network 阅读笔记相关推荐

High Performance Visual Tracking with Siamese Region Proposal Network全文翻译
摘要近年来,视觉对象跟踪一直是一个基本主题,许多基于深度学习的跟踪器在多个基准测试中取得了最先进的性能.然而,这些跟踪器中的大多数很难以实时速度获得最佳性能.在本文中,我们提出了 Siamese ...
High Performance Visual Tracking with Siamese Region Proposal Network 论文学习
文章目录论文阅读总结 Translation Abstract 1 Introduction 2 Related Works 2.1 Trackers based on Siamese networ ...
走进VOT--《High Performance Visual Tracking with Siamese Region Proposal Network》阅读翻译
前言:siamRPN是Siamfc之后的又一突破.SiamFC的缺点: Siamese的方法只能得到目标的中心位置,但是得不到目标的尺寸,所以只能采取简单的多尺度加回归,这即增加了计算量,同时也不够精 ...
CVPR 2018 Siam-RPN:《High Performance Visual Tracking with Siamese Region Proposal Network》论文笔记
理解出错之处望不吝指正. 本文模型叫做Siam-RPN.本文将Siamese Network和RPN结合,提出了一种端到端的离线训练方法,并把tracking过程视为one-shot detectio ...
论文阅读：Saliency-Guided Region Proposal Network for CNN Based Object Detection
论文阅读:Saliency-Guided Region Proposal Network for CNN Based Object Detection (1)Author (2)Abstract (3 ...
目标检测方法简介:RPN(Region Proposal Network) and SSD(Single Shot MultiBox Detector)
原文引用:http://lufo.me/2016/10/detection/ 最近几年深度学习在计算机视觉领域取得了巨大的成功,而在目标检测这一计算机视觉的经典问题上直到去年(2015)才有了完全使用 ...
RPN(Region Proposal Network)
RPN(Region Proposal Network) 学习RPN前最好先过一遍RCNN和Fast RCNN,本文的图来自原论文和bvBV1af4y1m7iL,有纰漏之处欢迎在评论区指出 RPN什么 ...
RPN（Region Proposal Network）提取候选框
前言 RPN全称是Region Proposal Network,也可理解为区域生成网络,或区域候选网络:它是用来提取候选框的. 目录一.RPN的由来二.RPN思路流程三.feature map ...
Detecting Visual Relationships with Deep Relational Networks（阅读笔记）
Detecting Visual Relationships with Deep Relational Networks(阅读笔记) 原文链接:https://blog.csdn.net/xue_we ...
DO DIFFERENT TRACKING TASKS REQUIRE DIFFERENT APPEARANCE MODELS?——阅读笔记
<DO DIFFERENT TRACKING TASKS REQUIRE DIFFERENT APPEARANCE MODELS?>--阅读笔记 Paper:https://arxiv.o ...

High Performance Visual Tracking with Siamese Region Proposal Network 阅读笔记

High Performance Visual Tracking with Siamese Region Proposal Network 阅读笔记相关推荐

最新文章

热门文章