gvnn: Neural Network Library for Geometric Computer Vision
| Autor: | Viorica Patraucean, John McCormac, Andrew J. Davison, Michael Bloesch, Simon Stent, Ankur Handa |
|---|---|
| Přispěvatelé: | Hua, G, Jegou, H, Dyson Technology Limited |
| Rok vydání: | 2016 |
| Předmět: |
FOS: Computer and information sciences
Technology Computer science Computer Vision and Pattern Recognition (cs.CV) Spatial transformer networks Computer Science - Computer Vision and Pattern Recognition 02 engineering and technology Unsupervised learning Computer Science Artificial Intelligence Machine Learning (cs.LG) Geometric networks Geometric vision Computer Science Theory & Methods Cellular neural network 0202 electrical engineering electronic engineering information engineering Artificial Intelligence & Image Processing Computer vision Image warping Visual odometry Transformer (machine learning model) Parametric statistics 08 Information And Computing Sciences Science & Technology Computer Science Information Systems Artificial neural network business.industry Deep learning Geometric transformation 020207 software engineering Backpropagation Computer Science - Learning Computer Science::Computer Vision and Pattern Recognition Computer Science 020201 artificial intelligence & image processing Artificial intelligence business |
| Zdroj: | Lecture Notes in Computer Science ISBN: 9783319494081 ECCV Workshops (3) 14th European Conference on Computer Vision (ECCV) |
| DOI: | 10.1007/978-3-319-49409-8_9 |
| Popis: | We introduce gvnn, a neural network library in Torch aimed towards bridging the gap between classic geometric computer vision and deep learning. Inspired by the recent success of Spatial Transformer Networks, we propose several new layers which are often used as parametric transformations on the data in geometric computer vision. These layers can be inserted within a neural network much in the spirit of the original spatial transformers and allow backpropagation to enable end-to-end learning of a network involving any domain knowledge in geometric computer vision. This opens up applications in learning invariance to 3D geometric transformation for place recognition, end-to-end visual odometry, depth estimation and unsupervised learning through warping with a parametric transformation for image reconstruction error. Comment: Submitted to ECCV Workshop on Deep Geometry |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|