In September, I will be start­ing a law de­gree at UBC’s Allard School of Law. My in­ter­ests in this area are broad, and I ex­pect they may change, but at the out­set, I want to fo­cus on the parts of law that help give the weak­est of us a voice, in the jus­tice sys­tem and in so­ci­ety in gen­er­al. This in­cludes de­fen­dant rights, free ex­pres­sion, and in some re­spects, copy­right. I also want to help with pub­lic le­gal ed­u­ca­tion and ac­cess-to-jus­tice ini­tia­tives, to help give more of us more confidence in the out­comes of the jus­tice sys­tem and oth­er gov­ern­ment de­ci­sion-mak­ing process­es.

I am trained as a com­put­er sci­en­tist; in that field, I’ve fo­cused on vi­su­al ob­ject recog­ni­tion, ma­chine learn­ing, and ap­pli­ca­tions of deep neur­al net­works.

Currently: data sci­en­tist at Kobo, vol­un­teer with the BCCLA, am­a­teur sprint­er.

Formerly: Head of R&D for Shelfie, Class 1 Flight Instructor (still cur­rent but not ac­tive), Canadian Forces Officer (Cadet Instructor Cadre), 2× Google Intern.

Selected Publications

Sancho McCann. “Object classification and lo­cal­iza­tion us­ing spa­tial­ly lo­cal­ized fea­tures”. Ph.D. Dissertation. UBC Department of Computer Science. 2014. [pdf]

Sancho McCann and David G. Lowe. “Spatially Local Coding for Object Recognition.” ACCV, 2012. [pdf] [poster] [project page]

Sancho McCann and David G. Lowe. “Local Naive Bayes Nearest Neighbor for Image Classification.” CVPR, 2012. [pdf] [project page]

A more com­plete list is at my Google Scholar profile.

Selected Projects

AtmosView: Visualization Redesign

I cre­at­ed Atmos­View, a new vi­su­al­iza­tion of at­mos­pher­ic sound­ing data (ver­ti­cal profiles of the at­mos­phere’s tem­per­a­ture and hu­mid­i­ty). People use this data to pre­dict soar­ing con­di­tions, at­mos­pher­ic sta­bil­i­ty, and the like­li­hood of se­vere weath­er. Previous di­a­grams have been called the most difficult at­mos­pher­ic di­a­grams to read. Atmos­View helps peo­ple to see bet­ter the in­for­ma­tion they’re in­ter­est­ed in and al­lows for eas­i­er com­par­isons be­tween mul­ti­ple charts.

Humanoid Robot

I worked with Dr. Jacky Baltes to build a small-size hu­manoid ro­bot. I cod­ed in C and cross com­piled for the ARM proces­sor on a Sonqy Clie. I pro­grammed it to walk and to find and kick a ball. This was our en­try in the 2005 FIRA RoboWorld Cup.

Robot Airplane

At the University of Manitoba, I was part of a team that built a ro­bot air­plane that could take-off, fly a search pat­tern, and land—all auto­nom­ous­ly. The air­plane sent a video feed and teleme­try to a ground sta­tion, where one of our team­mates could mark tar­gets of in­ter­est and re­port their co­or­di­nates. We placed first out of sev­en­teen teams in a com­pe­ti­tion that in­clud­ed BYU, University of Texas, Cornell, MIT, and UCSD. I wrote much of the com­put­er vi­sion code, which trans­formed the video feed’s pix­el co­or­di­nates into GPS co­or­di­nates, and pre­sent­ed that in­for­ma­tion to our ground sta­tion’s op­er­a­tor.