These are the main areas of research I worked on before turning my attention to teaching full time. While I still tinker with image processing every now and then, most of my work these days goes toward my courses. Please note that I do not lead a graduate research group, nor do I supervise graduate students or postdocs.

Image Blending for Long Exposures from Video

With careful image blending, it is possible to generate arbitrarily long exposures even in bright daylight from video or ima sequences. The process allows for selective enhancement of moving structure, and can be used to render long-exposure video: video in which each frame appears to be a long exposure.

Pre-print of the paper presented at the Colour and Photometry in Computer Vision workshop @ ECCV 2012.

arXiv draft of a larger article describing the enhancement of moving structure.

Video Results for the arXiv draft.

Stochastic Image Denoising

This was a fairly fast bling image denoising method based on random walks - at each pixel, a random walk process is used to approximate a blur kernel for similar pixles in the immediate neighbourhood. The appearance of these pixels is then combined with weights given by the blur kernel to reduce noise at the central pixel. Easily paralelizable, but not competitive with current methods (it was back in 2009, but it's been a long time!)

Pre-print of the paper presented at the British Machine Vision Conference (BMVC) 2009.

Old binary implementation for Linux and Windows.

Perceptual Grouping and Image Segmentation

During my Ph.D. and post-doctoral work I focused on low-level image segmentation and perceptual grouping. This included various algorithms for grouping contour fragments into shapes, for segmenting images based on low-level appearance cues, and for grrouping keypoint descriptors by appearance to aid in recognition. For more information on these methods, please have a look at the publications list below.

Publications

F. J. Estrada, “Practical Robotics in Computer Science Using the Lego NXT: An Experience Report”, Conference on Innovation and Technology in Computer Science Education, Bologna, 2017

F. J. Estrada, A. Tafliovich, “Bridging the Gap Between Desired and Actual Qualifications of Teaching Assistants: An Experience Report”, Conference on Innovation and Technology in Computer Science Education, Bologna, 2017

F. J. Estrada, “Time-Lapse Fusion”, Colour and Photometry in Computer Vision workshop at ECCV , Florence, Italy, 2012

F. J. Estrada, D. J. Fleet, and A. D. Jepson, “ Stochastic Image Denoising”, British Machine Vision Conference, London, Sep. 2009

F. J. Estrada, P. Fua, V. Lepetit, and , S. Süsstrunk, “Appearance Based Keypoint Clustering”, IEEE International Conference on Computer Vision and Pattern Recognition, Miami, U.S.A., 2009.

F. J. Estrada, and A. D. Jepson, “Benchmarking Image Segmentation Algorithms”, International Journal of Computer Vision, Vol. 85, No. 2, p. 167, 2009

C. Fredembach, F. Estrada, and S. Süsstrunk, “Memory Colour Segmentation and Classification Using Class-specific Eigenregions”, Journal of the Society for Information Display, 2009

R. Achanta, S. Hemami, F. Estrada, and S. Süsstrunk, “Frequency-tuned Salient Region Detection”, IEEE International Conference on Computer Vision and Pattern Recognition, Miami, U.S.A., 2009

P. Denis, J. Elder, and F. Estrada, “Efficient Edge-Based Methods for Estimating Manhattan Frames in Urban Imagery”, European Conference on Computer Vision, part II, pp. 197-210, 2008

R. Achanta, F. Estrada, P. Wils, and S. Süsstrunk, “Salient Region Detection and Segmentation”, International Conference on Computer Vision Systems, pp. 66-75, 2008

F. J. Estrada, and A. D. Jepson, “Robust Boundary Detection With Adaptive Grouping”, 5th IEEE Workshop on Perceptual Organization in Computer Vision, NY, USA, 2006

F. J. Estrada, and J. H. Elder, “Multiscale Contour Extraction Based on Natural Scene Statistics”, 5th IEEE Workshop on Perceptual Organization in Computer Vision, NY, USA, 2006

F. J. Estrada, and A. D. Jepson, “Quantitative Evaluation of a Novel Image Segmentation Algorithm”, IEEE International Conference on Computer Vision and Pattern Recognition, San Diego, U. S. A., 2005

F. J. Estrada, and A. D. Jepson, “ Perceptual Grouping for Contour Extraction”, International Conference on Pattern Recognition, Cambridge, U. K., 2004

F. J. Estrada, A. D. Jepson, and C. Chennubhotla “Spectral Embedding and Min-Cut for Image Segmentation”, British Machine Vision Conference, London, U. K., 2004

Ph.D. thesis “Advances in Computational Image Segmentation and Perceptual Grouping”, Department of Computer Science, University of Toronto, Jun. 2005.