Datasets and Code : DOWNLOAD (see README.txt to play with the code.)
1. >> RF_Demo;
2. >> Img_reconst_Demo;
Software Licence: GNU LESSER GENERAL PUBLIC LICENSE V3.0
Hanchen Xiong's Homepage
Learning Early Visual Features with Diverse RBMs
Datasets and Code : DOWNLOAD (see README.txt to play with the code.)
1. >> RF_Demo;
2. >> Img_reconst_Demo;
Software Licence: GNU LESSER GENERAL PUBLIC LICENSE V3.0
Datasets and Code : DOWNLOAD (see README.txt to play with the code.)
1. >> RF_Demo;
2. >> Img_reconst_Demo;
Software Licence: GNU LESSER GENERAL PUBLIC LICENSE V3.0
A Demo Code on a synthetic corpus: DOWNLOAD
| synthetic corpus |
estimated topices |
 |
 |
Licence: GNU LESSER GENERAL PUBLIC LICENSE V3.0
Homogeneity
Analysis for Object-Action Relation Learning
1. Datasets:
A collection of objects assoicated with features and actions
DOWNLOAD
2. Object-Action Learning with Homogeneity Analysis : DOWNLOAD
Software Licence: GNU LESSER GENERAL PUBLIC LICENSE V3.0
Note: please cite following references when you use our data or code
@inproceedings{xiong2013homogeneity,
title={Homogeneity analysis for object-action relation reasoning in kitchen scenarios},
author={Xiong, Hanchen and Szedmak, Sandor and Piater, Justus},
booktitle={IJCAI Workshop on Machine Learning for Interactive Systems: Bridging the Gap Between Perception, Action and Communication},
pages={37--44},
year={2013},
organization={ACM}
}
1. Datasets:
A collection of objects assoicated with features and actions
DOWNLOAD
2. Object-Action Learning with Homogeneity Analysis : DOWNLOAD
Software Licence: GNU LESSER GENERAL PUBLIC LICENSE V3.0
Note: please cite following references when you use our data or code
@inproceedings{xiong2013homogeneity,
title={Homogeneity analysis for object-action relation reasoning in kitchen scenarios},
author={Xiong, Hanchen and Szedmak, Sandor and Piater, Justus},
booktitle={IJCAI Workshop on Machine Learning for Interactive Systems: Bridging the Gap Between Perception, Action and Communication},
pages={37--44},
year={2013},
organization={ACM}
}
3D Point Cloud Registration Datasets and Benchmark Code
1. Datasets:
Our 3D point cloud datasets are modified format of 3D objects/shapse downloaded from KIT database ( available http://i61p109.ira.uka.de/ObjectModelsWebUI/). For each object,
different sizes point clouds are generated (100,200, 500, 1000, 2000) to evalute the efficiency of algorithms.
DOWNLOAD
2. Benchmark Code:
In our experiments, we compare some popular 3D registration algorithms, e.g. ICP, Gaussian mixture, SoftAssign. We tested them under two circumstances: first, with large motions
second, with large amount of outliers. Some examples are shown here:
The whole implementation of benchmark framework and all algorithms can be downloaded here: DOWNLOAD.
Software Licence: GNU LESSER GENERAL PUBLIC LICENSE V3.0
Note: please cite following references when you use our data or code
@inproceedings{Xiong2013CRV,
title={Efficient, General Point Cloud Registration With Kernel Feature Maps},
author={Xiong, Hanchen and Szedmak, Sandor and Piater, Justus},
booktitle={International Conference on Computer and Robot Vision (CRV)},
pages={83--90},
year={2013},
organization={IEEE}
}