Abstract | ||
---|---|---|
Ensembles of randomized decision trees, known as Random Forests, have become a valuable machine learning tool for addressing many computer vision problems. Despite their popularity, few works have tried to exploit contextual and structural information in random forests in order to improve their performance. In this paper, we propose a simple and effective way to integrate contextual information in random forests, which is typically reflected in the structured output space of complex problems like semantic image labelling. Our paper has several contributions: We show how random forests can be augmented with structured label information and be used to deliver structured low-level predictions. The learning task is carried out by employing a novel split function evaluation criterion that exploits the joint distribution observed in the structured label space. This allows the forest to learn typical label transitions between object classes and avoid locally implausible label configurations. We provide two approaches for integrating the structured output predictions obtained at a local level from the forest into a concise, global, semantic labelling. We integrate our new ideas also in the Hough-forest framework with the view of exploiting contextual information at the classification level to improve the performance on the task of object detection. Finally, we provide experimental evidence for the effectiveness of our approach on different tasks: Semantic image labelling on the challenging MSRCv2 and CamVid databases, reconstruction of occluded handwritten Chinese characters on the Kaist database and pedestrian detection on the TU Darmstadt databases. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/TPAMI.2014.2315814 | Pattern Analysis and Machine Intelligence, IEEE Transactions |
Keywords | Field | DocType |
computer vision,decision trees,image classification,learning (artificial intelligence),object detection,CamVid database,Hough-forest framework,Kaist database,MSRCv2 database,TU Darmstadt database,classification level,complex problems,computer vision problems,contextual information integration,joint distribution,label transition learning,local level,locally implausible label configuration avoidance,machine learning tool,object classes,object detection,occluded handwritten Chinese character reconstruction,pedestrian detection,performance improvement,random forests,randomized decision tree ensembles,semantic image labelling,split function evaluation criterion,structural information,structured label information,structured label space,structured low-level predictions,structured output prediction integration,structured output space,Random forests,object detection,semantic image labelling,structured prediction | Computer vision,Decision tree,Object detection,Joint probability distribution,Computer science,Structured prediction,Exploit,Artificial intelligence,Random forest,Pedestrian detection,Machine learning,Semantics | Journal |
Volume | Issue | ISSN |
36 | 10 | 0162-8828 |
Citations | PageRank | References |
23 | 0.90 | 44 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Peter Kontschieder | 1 | 376 | 21.10 |
Rota Bulo, S. | 2 | 72 | 3.70 |
Marcello Pelillo | 3 | 1888 | 150.33 |
Horst Bischof | 4 | 8751 | 541.43 |