Name
Abstract | ||
|---|---|---|
The choice of an appropriate representation is important when reconstructing motion from images. In this letter we propose a new type of deep neural network (DNN) that maps images to spatial paths represented by a recently introduced motion representation called arc-length dynamic movement primitive (AL-DMP). This representation separates the spatial from temporal aspects of motion and is therefore suitable for processing data that do not contain temporal information. We propose a physically meaningful loss function for training of AL-DMPs, which improves the performance of the trained DNN, and derive its gradients, which are needed to apply the backpropagation algorithm. Moreover, the proposed DNN architecture supports the processing of variable-size input images and images with cluttered background. The developed approach was applied to the reproduction of handwritten digits from single images that do not contain temporal aspects of motion. Our experiments also show that the proposed network can be applied to input images of sizes that are different from the size of training images. Finally, the proposed approach was successfully applied for reproducing real handwritten digits with a humanoid robot. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1109/LRA.2020.3039937 | IEEE Robotics and Automation Letters |
Keywords | DocType | Volume |
Deep learning for visual perception,novel deep learning methods,perception-action coupling | Journal | 6 |
Issue | ISSN | Citations |
1 | 2377-3766 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Rok Pahic | 1 | 1 | 2.04 |
| Andrej Gams | 2 | 385 | 29.54 |
| Ales Ude | 3 | 898 | 85.11 |