Get short, timely messages from Jason Corso.

Twitter is a rich source of instantly updated information. It's easy to stay updated on an incredibly wide variety of topics. Join today and follow @ProfJasonCorso.

Get updates via SMS by texting follow ProfJasonCorso to 40404 in the United States
Codes for other countries

Country	Code	For customers of
Two-way (sending and receiving) short codes:
Australia	0198089488 Telstra
Canada	21212 (any)
United Kingdom	86444 Vodafone, Orange, 3, O2
Indonesia	89887 AXIS, 3, Telkomsel
Ireland	51210 O2
India	53000 Bharti Airtel, Videocon
Jordan	90903 Zain
New Zealand	8987 Vodafone, Telecom NZ
United States	40404 (any)

ProfJasonCorso

Fix to our CVPR 2012 Action Bank paper now posted (fixed some erroneous results on HMDB51 and UCF50): goo.gl/mtSGu #news 1:13 PM May 31st via web
Action Bank v1.0 software and data now publicly available: goo.gl/sQD3G direct code link: goo.gl/lHMsw #news 2:06 PM May 29th via web
2 papers accepted at ECDM 2012. 1 on max-margin metric learning and 1 on spectral active clustering. #news 6:48 AM May 29th via web
Our KDD 2012 paper on Random Forest Distance for metric learning is available: goo.gl/V2pVg #news 1:51 PM May 25th via web
Our AMDO 2012 paper on dynamic human pose for activity recognition is available: goo.gl/udHXB #news 3:42 AM May 18th via web
Our ICIP 2012 paper on dictionary transfer for image denoising is available: goo.gl/9HH7s #news 12:39 PM May 17th via web
Our RSS 2012 paper on estimating human dynamics from monocular video is available: goo.gl/HLdd9 #news 11:16 AM May 17th via web
I will be an Area Chair for CVPR 2013. #news 7:12 AM May 17th via web
LIBSVX v1.0 Released. A Supervoxel Library and Benchmark for Early Video Processing. See goo.gl/ou03e #news 7:54 PM May 9th via web
#news 5 new accepted papers: 1 KDD (metric learning), 1 ICIP (domain transfer), 1 RSS (human pose), 1 AMDO, 1 ISVC. Details to follow. 7:10 PM May 6th via web
#news New video of our work in modeling and localizing stairways for multi-floor exploration by autonomous robots. youtu.be/fGkA7srkgs8 10:31 AM Apr 18th via web
Article on a full-scene parts-based model/algorithm accepted at PRL SI on Scene Understanding. Preprint: goo.gl/Xy3f4 #news 6:35 AM Apr 12th via web
CVPR2012 camera ready papers available: action bank goo.gl/mtSGu and supervoxels goo.gl/pgqj1 (code/data coming soon) #news 1:05 PM Mar 27th via web
#news Two papers accepted at CVPR. cse.buffalo.edu/~jcorso/pubs/i… 6:51 AM Mar 1st via web
#news Release of our Python-based Video IO/processing library: nD-vision. launchpad.net/ndvision Checkout with: bzr branch lp:ndvision 1:58 PM Dec 20th, 2011 via web
New tech-report on some work in high-level scene understanding with a parts-based model. goo.gl/mUupD arxiv.org/abs/1108.4079 #news 7:49 PM Aug 22nd, 2011 via web
My poster "Layered Models for Bridging from Low to High Level Vision" for the Frontiers in CV Workshop available: goo.gl/Hs99k #news 11:12 PM Aug 19th, 2011 via web
New tech report on some work using ideas from dynamical systems to extract features for face detection: arxiv.org/abs/1108.3525 #news 5:50 PM Aug 17th, 2011 via web
#news Some local news coverage on our DURIP grant. goo.gl/zEF8m 6:23 PM Aug 10th, 2011 via web
@Dariolakis Ah, thanks! Indeed many of the ideas in that work build off of what we did back then! 6:22 AM Jul 17th, 2011 via web

Name Jason Corso
Location Buffalo, NY
Web http://www.cse.bu...
Bio Corso is an asst. professor of computer science at SUNY Buffalo. He works in the areas of computer vision, semantic computing, and machine learning.