In this demo we propose an automated flow that allows the transparent execution of ordinary code on a heterogeneous platform including an FPGA. Our solution requires no change in the code, not even pragma indications to guide the optimization, and dynamically adapts its behaviour to the available data and the workload of the system. Thus, the developer does not need to be aware of the target platform details, nor she has to forecast usage patterns to prevent performance bottlenecks, as the system transparently identifies parallelizable, computationally-intensive code fragments and dispatches them to a data flow overlay architecture built on top of the FPGA. Since the bitstream we use is fixed, and contrary to HLS, we can alter the functionalities offered by the FPGA on-the-fly to adapt them to current usage. Finally, since we operate at the LLVM's Intermediate Representation (IR) level, our approach is language-agnostic.

Learning filters to produce sparse image representations in terms of over-complete dictionaries has emerged as a powerful way to create image features for many different purposes. Unfortunately, these filters are usually both numerous and non-separable, making their use computationally expensive. In this paper, we show that such filters can be computed as linear combinations of a smaller number of separable ones, thus greatly reducing the computational complexity at no cost in terms of performance. This makes filter learning approaches practical even for large images or 3D volumes, and we show that we significantly outperform state-of-the-art methods on the curvilinear structure extraction task, in terms of both accuracy and speed. Moreover, our approach is general and can be used on generic convolutional filter banks to reduce the complexity of the feature extraction step.

Learning filters to produce sparse image representations in terms of over complete dictionaries has emerged as a powerful way to create image features for many different purposes. Unfortunately, these filters are usually both numerous and non-separable, making their use computationally expensive. In this paper, we show that such filters can be computed as linear combinations of a smaller number of separable ones, thus greatly reducing the computational complexity at no cost in terms of performance. This makes filter learning approaches practical even for large images or 3D volumes, and we show that we significantly outperform state-of-the-art methods on the linear structure extraction task, in terms of both accuracy and speed. Moreover, our approach is general and can be used on generic filter banks to reduce the complexity of the convolutions.

Recent years have seen an increasing interest in sparse representations for image classification and object recognition, probably motivated by evidence from the analysis of the primate visual cortex. It is still unclear, however, whether or not sparsity helps classification. In this paper we evaluate its impact on the recognition rate using a shallow modular architecture, adopting both standard filter banks and filter banks learned in an unsupervised way. In our experiments on the CIFAR-10 and on the Caltech-101 datasets, enforcing sparsity constraints actually does not improve recognition performance. This has an important practical impact in image descriptor design, as enforcing these constraints can have a heavy computational cost.