Comments (4)
HonglinChu
commented on May 26, 2024
你是模型加载错了吧。alexnet的那个模型建议自己重新训练一下，我自己训练的没有调参，在otb测试很好，但是在vot18上一般，比论文低了两个点。
…------------------ 原始邮件 ------------------
发件人: "zhangfangbin"<[email protected]>;
发送时间: 2020年12月12日(星期六) 中午12:21
收件人: "HonglinChu/SiamTrackers"<[email protected]>;
抄送: "Subscribed"<[email protected]>;
主题: [HonglinChu/SiamTrackers] 关于siamfc (#23)
用alexnet训练后，在otb和vot2018上进行测试，然后改成resnet22训练，再在otb和vot2018上测试，得到的结果没有变，一个数字都没变，这是为什么。
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or unsubscribe.
from siamtrackers.
Rongmiq
commented on May 26, 2024
class SiamFCTransforms(object):
def __init__(self, exemplar_sz=127, instance_sz=255, context=0.5):
self.exemplar_sz = exemplar_sz
self.instance_sz = instance_sz
self.context = context
self.transforms_z = Compose([
RandomStretch(),
CenterCrop(instance_sz - 8), #255-8
RandomCrop(instance_sz - 2 * 8), #255-16
CenterCrop(exemplar_sz), #127
ToTensor()])
self.transforms_x = Compose([
RandomStretch(),
CenterCrop(instance_sz - 8), # 247
RandomCrop(instance_sz - 2 * 8), #239
ToTensor()])
请问fc在训练时对于搜索分支x为什么不crop成 255x255 的，而是使用 239x239 的大小，这也导致搜索分支输出20x20 以及最后的得分图大小不是 17x17，而是15x15？
from siamtrackers.
HonglinChu
commented on May 26, 2024
这个你也可以改成255x255的试试，instance-sz是255大小，随机裁剪之就是255-8*2=239。我的理解是这里主要是为了进一步增强泛化能力。你也可以不进行这个操作进行对比，毕竟是非官方代码，不一定和原文一致的。
from siamtrackers.
HonglinChu
commented on May 26, 2024
用alexnet训练后，在otb和vot2018上进行测试，然后改成resnet22训练，再在otb和vot2018上测试，得到的结果没有变，一个数字都没变，这是为什么。
from siamtrackers.
Related Issues (20)
- 为什么输入的两个图像是127/255，这两个超参数有什么解释吗？
- par_crop的问题 HOT 1
- 关于有遮挡数据的训练 HOT 1
- 骨干网络的训练 HOT 1
- siamban导入toolkit.utils.region 报错
- 关于Nanotrack特征融合的部分 HOT 2
- 如图，SiamFCpp-pysot和SiamFCpp-video_analyst都是SiamFC++的实现吗，具体模型参数是否一样，有没有区别 HOT 1
- nanotrack数据集got裁剪的问题 HOT 2
- nanotrack v1复现问题 HOT 3
- NanoTrack为什么要将训练数据裁剪为271而不是pysot通用的511？ HOT 3
- nanotrack V3复现过程多数据集训练和got10训练效果一样？ HOT 3
- 请问大家的nanotrack V3最多能练到多少？ HOT 25
- siamFC预训练模型 HOT 1
- NanoTrack train on custom dataset HOT 1
- NanoTrack自定义数据集相关问题 HOT 2
- YTB&VID 数据好像失效了，麻烦能更新一下数据集吗 HOT 1
- onnx HOT 1
- Cant train good! HOT 2
- VOT2018.json在哪里啊
- Readme里面的数据，为啥和我在论文中看到的不一致 HOT 4
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from siamtrackers.