A. Barbu, G.Gramajo. Face Detection Using a 3D Model on Face Keypoints. (arxiv)
J. Anaya, A. Barbu. RENOIR - A Benchmark Dataset for Real Noise Reduction Evaluation. (pdf)
A. Barbu, Y. She, L. Ding, G. Gramajo. Feature Selection with Annealing for Big Data Learning. (arxiv)
A. Barbu, T.F. Wu, Y. N. Wu. Learning Mixtures of Bernoulli Templates by Two-Round EM with Performance Guarantee. Accepted in Electronic Journal of Statistics (arxiv)
L. Ding, A. Barbu. Scalable Subspace Clustering with Application to Motion Segmentation. Current Trends in Bayesian Methodology with Applications. S.K. Upadhyay. D.K. Dey, U. Singh, A. Loganathan, Editors. Chapman&Hall/CRC Press. 2014 (pdf)
A. Barbu, M. Pavlovskaia, S.C. Zhu. Rates for Inductive Learning of Compositional Models. AAAI Workshop Replearn 2013 (pdf)
A. Barbu. Multi-Path Marginal Space Learning for Object Detection. Academic Press Library in Signal Processing: Volume 4: Image, Video Processing and Analysis, Hardware, Audio, Acoustic and Speech Processing. pp 271–291, 2013 (pdf)
A. Barbu. Hierarchical Object Parsing from Structured Noisy Point Clouds. IEEE PAMI, 35, No. 7, 1649–1659, 2013. (pdf, arxiv, slides)
L. Ding, A. Barbu, A. Meyer-Baese. Learning a Quality-Based Ranking for Feature Point Trajectories. ACCV 2012 (pdf)
L. Ding, A. Barbu, A. Meyer-Baese. Motion Segmentation by Velocity Clustering with Estimation of Subspace Dimension. ACCV Workshop DTCE 2012 (pdf)
A. Barbu, N. Lay. An Introduction to Artificial Prediction Markets for Classification. Journal of Machine Learning Research, 13, 2177–2204, 2012. (pdf)
A. Barbu, M. Suehling, X. Xu, D. Liu, S. K. Zhou, D. Comaniciu. Automatic Detection and Segmentation of Lymph Nodes from CT Data. IEEE Trans Medical Imaging, 31, No. 2, 240–250, 2012.(pdf)
W. Wu, T. Chen , A. Barbu, P. Wang, N. Strobel, S. Zhou, D. Comaniciu. Learning-based Hypothesis Fusion for Robust Catheter Tracking in 2D X-ray Fluoroscopy. CVPR 2011 (pdf)
F. Bunea, A. Tsybakov, M. Wegkamp and A.Barbu. SPADES and mixture models. Annals of Statistics 38, No. 4, 2525–2558, 2010.(pdf)
F. Bunea and A.Barbu. Dimension reduction and variable selection in case control studies via regularized likelihood optimization. Electronic Journal of Statistics, 3, 2009. (pdf)
A. Barbu. Training an Active Random Field for Real-Time Image Denoising. IEEE Trans. Image Processing, 18, November 2009. (pdf, ppt)
S. Seifert, A. Barbu, S. Zhou, D. Liu, J. Feulner, M. Huber, M. Suehling, A. Cavallaro, D. Comaniciu. Hierarchical parsing and semantic navigation of full body CT data. SPIE Medical Imaging, 2009 (pdf)
Y. Zheng, A. Barbu, B. Georgescu, M. Scheuering and D. Comaniciu. Four-Chamber Heart Modeling and Automatic Segmentation
for 3D Cardiac CT Volumes Using Marginal Space Learning and Steerable Features. IEEE Trans Medical Imaging, November 2008. (pdf)
L. Lu, A. Barbu, M. Wolf, J. Liang, M. Salganicoff, D. Comaniciu. Accurate Polyp Segmentation for 3D CT Colonography Using Multi-Staged Probabilistic Binary Learning and Compositional Model. CVPR 2008.(pdf)
A. Barbu, V. Athitsos, B. Georgescu, S. Boehm, P. Durlak, D. Comaniciu. Hierarchical Learning of Curves: Application to Guidewire Localization in Fluoroscopy. CVPR 2007 (pdf)
A. Barbu, S.C. Zhu. Generalizing Swendsen-Wang for Image Analysis. J. Comp. Graph. Stat. 16, No 4, 2007 (pdf)
A. Barbu, S.C. Zhu. Generalizing Swendsen-Wang to sampling arbitrary posterior probabilities, PAMI, 27, August 2005 (pdf)
A. Barbu, S.C. Zhu. Multigrid and Multi-level Swendsen-Wang Cuts for Hierarchic Graph Partition, CVPR 2004 (pdf)

Search Google Scholar for Adrian Barbu OpenAlex Works Search

Memberships

IEEE, Member; 2004 - present

Technologies

System and method for Generating a Benchmark Dataset for Real noise Reduction Evaluation

Often images taken with smartphones or point-and-shoot digital cameras come out noisy due to lack of sufficient lighting. This low-light noise problem is widespread, being present in all of the smartphones in the world, more than 1 billion total. This problem generates in consumers disappointment and frustration with the quality of the images taken in low light. While a number of commercial denoising packages are already available on the market, the majority of them are trained on images corrupted by artificial noise, rather than trained on real low-light noisy images. Since artificial noise has different characteristics than real noise, these packages do not perform as well as a denoising algorithm trained on images corrupted by real noise as we have created.

We have developed a fully automatic state of the art algorithm (RENOIR) for denoising smartphone and digital camera images which have low-light noise problems. The RENOIR algorithm could be either; be sold directly to the public as a standalone application or could be licensed to smartphone or digital camera manufacturers to be embedded in their devices.

CNN Filters for Noise Estimation and Improved Denoising in Low-Light Noisy Images

The proposed invention is a system and method for training a Convolutional Neural Network (CNN) to predict a tuning parameter to be used in an existing image denoising method (called BM3D) to obtain best possible denoising results on images obtained by digital cameras in low-light conditions. The performance of the BM3D denoising algorithm varies with this tuning parameter.

In this work we present a method to predict the best parameter value for each image patch and we observe that using this prediction we obtain better results than using a fixed parameter value for all images.

There are many image denoising methods available today. However, they are trained and tested on artificial noise. According to our observations, when it comes to images corrupted by real low light noise the BM3D method works best. Our work takes the BM3D method and enhances it by predicting what its tuning parameter should be for each image patch being denoised.

This technology could be directly sold to consumers in the form of an app or embedded in a mobile phone or digital camera.