Comments (13)
@johnSmith1990 You can use any of these projects to train Yolov3.
https://github.com/ultralytics/yolov3
https://github.com/AlexeyAB/darknet
BTW are you able to detect+track successfully?
from detection_tracking_jetsontx2.
Mashkorin @MuhammadAsadJaved,
the detection model is tiny here? or full yolo? cfg file seems to be for tiny yolo.
I am trying to run detect and track but I have to compile caffe first. But in general I am working on tracking project.
I will share my results here.
from detection_tracking_jetsontx2.
Mashkorin @MuhammadAsadJaved,
the detection model is tiny here? or full yolo? cfg file seems to be for tiny yolo.I am trying to run detect and track but I have to compile caffe first. But in general I am working on tracking project.
I will share my results here.
You are right. It's Yolov3-Tiny.
You can read this issue we are facing from several days. I also mentioned the caffe build process in the issue.
The detector is detecting very well, You can try with python3 mainDetect.py
But the tracker and tracker combined with Yolo is not working. I hope you will debug the code.
You can try only tracker with python3 mainGoturnTracker.py
from detection_tracking_jetsontx2.
thanks.
First of all, I tested several video for detection but detection failed in many frames. so I am trying to test full yolo for good FPS results. Then train for drone detection. Then compiling caffe and test tracker. This is my roadmap.
I want to run my project on 1080 ti.
Do you use full yolo or tiny version? Is full yolo real time?
from detection_tracking_jetsontx2.
thanks.
First of all, I tested several video for detection but detection failed in many frames. so I am trying to test full yolo for good FPS results. Then train for drone detection. Then compiling caffe and test tracker. This is my roadmap.
I want to run my project on 1080 ti.Do you use full yolo or tiny version? Is full yolo real time?
I am using same given weights.
I think first you should try to use the original code and make sure it can work or not, if it can then you can train yolov3 to replace detector. Otherwise it will waste your time to train if tracker can not work. So my suggestion is first we should make tracker work and then train detector.
I just figured it out the problem is with the tracker draw function.
https://github.com/tau-adl/Detection_Tracking_JetsonTX2/blob/master/DetectAndTrackUtils.py#L58
I just checked the x1 and x2 , y1 and y2 values was same, so tracker was not drawing bounding box. So i just added 50 and 30 in x2 and y2 to see the effect of the bounding box. Now it's drawing the bounding box but just after the detection ,the tracker's bounding box start moving at the bottom left. as you can see the first picture is detection and the second picture is very next frame after detection when it start tracking. So now I want to figure it out that why tracker's draw function is getting wrong values for bounding box.
from detection_tracking_jetsontx2.
Ok.
from detection_tracking_jetsontx2.
@MuhammadAsadJaved May you help me in compiling caffe? I am using python 3.7 cuda 10,windos 10 OS.
from detection_tracking_jetsontx2.
@johnSmith1990 So sorry I never tried in windows os.
from detection_tracking_jetsontx2.
@MuhammadAsadJaved Hello
How are you?
Do you understand how are the annotation of Alov300++ dataset?
I want to label my videos and train goturn network.
regards
from detection_tracking_jetsontx2.
from detection_tracking_jetsontx2.
Hi @MuhammadAsadJaved.
I got the goturn for pytorch and run it successfully(https://github.com/nrupatunga/goturn-pytorch) and I want to combine it with yolo network.
from detection_tracking_jetsontx2.
@johnSmith1990 Hello friends, have you successfully trained the tracking algorithm? Is the GOTURN algorithm used? Is the code given by the author? If it is, it means that there is no problem with the tracking algorithm code. The problem may appear in the pre- trained weight file of GOTURN.Because you successfully ran the GOTURN algorithm, and another friend @MuhammadAsadJaved successfully ran the yolov3 detection algorithm
from detection_tracking_jetsontx2.
Hi @gzc-123456-gzc
I run goturn successfully on my machine but because of problem in labeling of my data for goturn training also low FPS of goturn on gpu, i decided to pause using it. I switched and I am searching other methods to customize.
from detection_tracking_jetsontx2.
Related Issues (15)
- No module named 'videoUtils' HOT 2
- Expo884FPS20.mp4 test video HOT 7
- OpenCV Cuda enabled? HOT 5
- Yolov3 (for different classes) + GOTURN HOT 10
- Python version? HOT 2
- How to install torch on TX2? HOT 2
- No Detection or Tracking HOT 23
- Where are weights?? HOT 4
- Simultaneous detection and tracking
- Is there anyone able to track successfully? HOT 11
- YOLO training
- YOLO training HOT 1
- running on laptop HOT 1
- Which Jetson TX2 version have you used?
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 detection_tracking_jetsontx2.