Monthly Archives: April 2016

1st RECOD hands-on deep learning workshop

The first RECOD hands-on deep learning workshop took place on last April 16-17. The workshop was open to all RECODists (students and researchers) and the goal was to take a very practical approach over the basic models of deep learning. … Continue reading

Posted in blog, extra | Tagged , , , , | Leave a comment

Visualizing and Interacting with Kernelized Data

Kernel methods have emerged as a versatile mechanism for data classification, clustering and pattern recognition. Understanding of the mapping performed by a kernel simplifies the choice and the design of kernels as well as the fine-tuning of kernel parameters, thus … Continue reading

Posted in blog, publications, science | Tagged , , , , , , , , , , , , , | Leave a comment

RECOD paper has been selected to compose the 2015 list of honored papers in the Computers in Biology and Medicine journal

We are very happy to announce that our paper “Mid-level image representations for real-time heart view plane classification of echocardiograms” has been selected to compose the 2015 list of honored papers in the Computers in Biology and Medicine journal. The … Continue reading

Posted in awards, blog | Tagged , , , , , , , , , , , , , , | Leave a comment

A Manifold Learning Approach for Personalizing HRTFs from Anthropometric Features

This research paper presents a new anthropometry-based method to personalize head-related transfer functions (HRTFs) using manifold learning in both azimuth and elevation angles with a single nonlinear regression model. The authors, from Unicamp and Redmond Microsoft Research, propose a graph … Continue reading

Posted in blog, publications, science | Tagged , , , , , , , , , , , , , , , , , , , , | Leave a comment

Article is the second most cited in JVCI

The number of citations is one good metric, although not the only one, to evaluate the quality of a published research paper. We are very happy to highlight that our article, entitled “Comparative study of global color and texture descriptors … Continue reading

Posted in awards, blog | Tagged , , , , | Leave a comment

Multi-directional and Multi-scale Perturbation Approaches for Blind Forensic Median Filtering Detection

In this work, the Ph.D. candidate Anselmo Ferreira along with professors Jefersson dos Santos (UFMG) and Anderson Rocha (RECOD) propose multi-scale and multi-directional median filtering detection algorithms to identify the presence of image tampering in digital images. These algorithms are … Continue reading

Posted in blog, publications, science | Tagged , , , , , , | Leave a comment

RECOD partnership with Samsung on the news

Recent news published at Inova Unicamp (in Portuguese) highlights the efforts of RECOD towards the development of two systems: an echocardiogram image classifier, which is suitable for use in real-time setups; and an open-set classifier capable of dealing with multi-class … Continue reading

Posted in blog, media | Tagged , , , , , , , | Leave a comment

Talk: On the Fine-Tuning Parameters in Deep Boltzmann Machines Using Quaternions

In talk, which happened last Tuesday, March 29th, 2016, Prof. João Paulo Papa addressed how RBM (Restricted Boltzmann Machines) can be used in the Machine Learning field as an alternative or complement to CNNs. The talk also discussed the parameter … Continue reading

Posted in blog, science, talk | Tagged , , , , , , , | 2 Comments