Welcome to my personal page at cs.rit.edu. My name is Kenny Mauricio Davila Castellanos. I am a PhD candidate in
the program of Computing and Information Sciences at Rochester Institute of Technology. I am currently doing research
related to the fields of Computer Vision, Pattern Recognition and Information Retrieval. My thesis topic is retrieval of math formulas
from documents and
videos.
I am also interested as well in the more general field of
Artificial Intelligence,
and I also love the fields of
Computer Graphics and
Robotics.
I currently work at the
Document and Pattern Recognition Lab (DPRL) under the supervision
of Dr. Richard Zanibbi
Contact:
Email: kxd7282@rit.edu
LinkedIn: Kenny Davila
Publications
Google Scholar: Citations
K. Davila. "AppearanceBased Retrieval of Mathematical Notation in Documents and Lecture Videos". In Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval (SIGIR 2016). ACM.
R. Zanibbi, K. Davila, A. Kane, F.W. Tompa. "MultiStage Math Formula Search: Using AppearanceBased Similarity Metrics at Scale". In Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval (SIGIR 2016) . ACM.
H. Chatbri, K. Davila, K. Kameyama and R. Zanibbi. "Shape matching using keypoints extracted from both the foreground and the background of binary images". In International Conference on Image Processing Theory, Tools and Applications (IPTA 2015). IEEE.
K. Davila, S. Ludi and R. Zanibbi. "Using Offline Features and Synthetic Data for Online Handwritten Math Symbol Recognition". 14th International Conference on Frontiers in Handwriting Recognition (ICFHR 2014). IEEE.
K. Davila, A. Agarwal, R. Gaborski, R. Zanibbi and S. Ludi. "AccessMath: Indexing and Retrieving Video Segments Containing Math Expressions Based on Visual Similarity". In IEEE Western New York Image Processing Workshop (WNYIPW 2013). IEEE.
M.Sc. Project Report: Math Expression Retrieval Implemented Through Sketches (Updated: May 9, 2013)
Research Projects
AccessMath

Tangent

Math Symbol Recognizer

AccessMath: Whiteboard Content Extraction and Retrieval from Lecture Videos
AccessMath is a project originally conceived with the goal of helping students with low vision both during lectures and
outside of the classroom. However, the tools that we are producing will be helpful for other students as well.
A collection of lecture videos contains a large amount of information which is hard to retrieve without the proper
labels or tags. In AccessMath, we are working on methods that will extract and index the content of math lecture
videos with minimal human intervention. After indexing this data, AccessMath will provide methods for retrieval based
on the visual similarity between the provided query images and the whiteboard content stored in the index.


Source Video Frame 
Content Extracted 
There are different challenges involved in the process of whiteboard content extraction from the videos. After extraction,
local features are computed to describe the whiteboard content on the images. The retrieval method used is recognitionfree
which means that no optical character recognition (OCR) is performed and everything is matched based on visual similarity.
K. Davila. "AppearanceBased Retrieval of Mathematical Notation in Documents and Lecture Videos". In Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval (SIGIR 2016). ACM.
K. Davila, A. Agarwal, R. Gaborski, R. Zanibbi and S. Ludi. "AccessMath: Indexing and Retrieving Video Segments Containing Math Expressions Based on Visual Similarity". In IEEE Western New York Image Processing Workshop (WNYIPW 2013). IEEE.
Technical Report: MS Project Report (Updated: May 9, 2013)
Tangent 3: Math Search Engine
Tangent is a scalable math search engine originally developed at DPRL.
Tangent version 3 has been developed in collaboration with Frank W. Tompa and Andrew Kane from University of Waterloo.
The current version of the search engine uses a twostage retrieval method. On the first stage, a coreengine quickly finds
good candidate matches from large databases using pairs of symbols from a Symbol Layout Tree (SLT) used to represent each formula.
On the second stage, a reranker applies a finer matching method which is able to unify variables and has partial support for
wildcard expansion. My main contributions to this project are the reranking functionality and visualization tools.


Matching a query with wildcard and variable unification 
SLT Representation of a candidate match 
Source Code and Data: Tangent @ DPRL website
K. Davila. "AppearanceBased Retrieval of Mathematical Notation in Documents and Lecture Videos". In Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval (SIGIR 2016). ACM.
R. Zanibbi, K. Davila, A. Kane, F.W. Tompa. "MultiStage Math Formula Search: Using AppearanceBased Similarity Metrics at Scale". In Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval (SIGIR 2016) . ACM.