2013 Brazilian Conference on Intelligent Systems (BRACIS) (2013)
Oct. 19, 2013 to Oct. 24, 2013
In this work we address the problem of feature extraction for image object recognition. We propose a new, learned, feature descriptor for images, the convolutional sparse descriptor, which is based on recent advances in machine learning. It computes a spatial representation of the entire input image based on feature responses of local descriptors. The feature responses are calculated using a learned dictionary, which is learned using the sparse coding algorithm, instead of the vector quantization (VQ). Experiments on the benchmark CIFAR-10 show that our method outperforms several state-of-the-art algorithms.
Feature extraction, Dictionaries, Vectors, Convolutional codes, Encoding, Image coding, Object recognition
E. F. Carvalho and P. M. Engel, "Convolutional Sparse Feature Descriptor for Object Recognition in CIFAR-10," 2013 Brazilian Conference on Intelligent Systems (BRACIS), Fortaleza, Brazil, 2014, pp. 131-135.