Abstract | ||
---|---|---|
Random forests are popular classifiers for computer vision tasks such as image labeling or object detection. Learning random forests on large datasets, however, is computationally demanding. Slow learning impedes model selection and scientific research on image features. We present an open-source implementation that significantly accelerates both random forest learning and prediction for image labeling of RGB-D and RGB images on GPU when compared to an optimized multi-core CPU implementation. We further use the fast training to conduct hyper-parameter searches, which significantly improves on earlier results on the NYU depth v2 dataset. Our flexible implementation allows to experiment with novel features, such as height above ground, which further increases classification accuracy. curfil prediction runs in real time at VGA resolution on a mobile GPU and has been used as data term in multiple applications. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1007/978-3-319-29971-6_22 | Communications in Computer and Information Science |
Keywords | DocType | Volume |
Random forest,Computer vision,Image labeling,GPU,CUDA | Conference | 598 |
ISSN | Citations | PageRank |
1865-0929 | 0 | 0.34 |
References | Authors | |
16 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hannes Schulz | 1 | 55 | 10.82 |
Benedikt Waldvogel | 2 | 24 | 1.21 |
Rasha Sheikh | 3 | 3 | 1.40 |
Sven Behnke | 4 | 1672 | 181.84 |