Comments (2)
Hi, @Juggernaut93 . One of the difference between MOT15 and MOT17 is the label of the occlusion rate. Mot17 provides this label. However, mot15 doesn't. Here are some examples of detections labels:
MOT17: mot17/train/MOT17-02-DPM/gt/gt.txt
135,1,912,484,97,109,0,7,1
136,1,912,484,97,109,0,7,0.93878
137,1,912,484,97,109,0,7,0.86735
138,1,912,484,97,109,0,7,0.79592
139,1,912,484,97,109,0,7,0.72449
The last column indicates the occlusion rate.
MOT15: train/ADL-Rundle-6/gt/gt.txt
1,1,1703,385,157,339,1,-1,-1,-1
1,3,1293,455,83,213,1,-1,-1,-1
1,4,259,449,101,261,1,-1,-1,-1
1,5,1253,529,55,127,1,-1,-1,-1
2,1,1699,383,159,341,1,-1,-1,-1
2,3,1293,455,83,213,1,-1,-1,-1
However, we cannot find the the occlusion rate labels.
Because we only extract the center point's feature for each detected boxes in our network. We cannot use these completely occluded detected boxes to train our network. Our description in this paper may be somewhat ambiguous. We'll improve this. Thank you for your suggestions.
from sst.
@shijieS thanks for your answer. Yes, I think your explanation here is clearer, now I understand the point :)
from sst.
Related Issues (20)
- Issues on Track age
- All_similarity and all_iou
- ToPercentCoords
- Input questions about the Network Diagram
- Question about fps HOT 1
- Issues with training HOT 1
- I'm trying to understand the code
- UserWarning HOT 3
- Base network selection HOT 3
- Apply for License HOT 2
- The final affinity matrix HOT 1
- Incorrect ID assignments in the project
- Why the accuracy and speed of retrained model is not as good as the sst300_0712_83000.pth which is given by author?
- About FPS HOT 3
- Two important parameters
- Why DAN get high IDSW score?
- Get Bad results when retraining the DAN on MOT17?
- There is always "None processing" while training on MOT15 dataset HOT 1
- 运行到一半出错
- Understanding the Assembly Loss in the loss function
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 sst.