Comments (5)
Pedestrian 0_5_12b
is a man with a small girl in a pink coat. There was a mistake in the attributes which is now corrected. This should resolve your questions regarding the frame range for the pedestrian.
In case there are other errors in the attributes, I suggest visualizing the data to establish the correspondences between the pedestrians and their labels. All annotations and videos can be viewed using CVAT. Version 0.3.0 is the one we used for the original annotations, I am not sure the XML files will work with the latest CVAT.
Poses were computed for all bounding boxes, even partially or fully occluded. Poses are likely not very good for those. In the benchmark occlusion was only present for a small percentage of the data since we only took the observation part, therefore we used it as is. You may want to experiment with filtering pose information by occlusion tag.
from jaad.
Hi Iuliia,
Many thanks for your reply and for correcting the attributes! That's very kind of you!
Also thanks a lot for your comments on the pose data in the benchmark since we are particularly interested in this.
By the way, can I ask have you performed normalization for the pose data? I have observed some pose data and found the range is between 0 and 1. So I am wondering this.
Thanks a lot again!
Bests,
Xingchen
from jaad.
Hi,
I have tried to use CVAT to visualise your annotations. The version I am using is the latest one I think (link is here https://cvat.org/tasks). However, the visualisation results and the annotation XML file do not match (not sure if this is related to the version of CVAT. I haven't successfully installed CVAT 0.3.0 yet).
The first problem is 'old_id' in the XML file is displayed instead of 'id'. For example, 'pedestrian' is displayed rather than '0_5_13b'.
The second problem is that, the displayed 'old_id' is not the same as that in the XML file. For example, the pedestrian '0_5_12b', its 'old_id' is pedestrian5, but in CVAT, 'PEDESTRIAN 7 (MANUAL)' is displayed.
This inconsistency exists for many other pedestrians...I am not sure if this is related to the version of CVAT, and don't know if you have encountered similar issues previously.
Bests,
Xingchen
from jaad.
I don’t recall having the issues you described with the latest cvat. Still, I do recommend using the older version in which the annotations were created.
from jaad.
OK, thank you very much for your quick reply!
I will try to install the older version again to see if that works.
Thanks!
Xingchen
from jaad.
Related Issues (18)
- How to Use the video annotation instead of images in tensorflow API HOT 1
- Any reason behind making intention = 0 when sample_type = 'all'? HOT 4
- steering angle of the vehicle HOT 1
- Normalization in pose data? HOT 2
- Using the interface to extract frames HOT 7
- annotations generation from .pkl file HOT 4
- av_interleaved_write_frame(): Input/output error HOT 2
- Sequence analysis for frames in a video HOT 1
- Vehicle odometry HOT 1
- Is it possible to add pedestrian_unique_id on retinanet method results? HOT 1
- Pedestrian tracking error HOT 1
- Calibration for camera available? HOT 2
- Purpose of new python files HOT 2
- Not able to use .mat file HOT 2
- Clarity on action and reaction HOT 14
- data format related problems HOT 1
- Pedestrian behavior annotations do not match in xmls HOT 2
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 jaad.