
Computer Vision and Machine Learning Research Manager @Dataminr. Views expressed are my own.
New York, USA
Joined September 2012
- Tweets53
- Following948
- Followers179
- Likes13.9K
📢 Call for #ACMMM2026 Industry Expert Talks
We’re inviting industry leaders, researchers, and practitioners to share insights on cutting-edge multimedia trends, real-world applications, and deployment challenges.
📩 Apply by May 23, 2026: 2026.acmmm.org/site/call-ind…
Svebor Karaman retweeted
🚀Passionate about building human-centered AI systems? @Dataminr's HAI team is hiring Software Engineers to create AI-enabled tools that enhance human workflows. Join our collaborative and innovative team, turning applied research into reality! #AI #HCI dataminr.wd12.myworkdayjobs.…
PhD School Talk @acmmm2022 announcement: “The Importance of Iteration on the Path from Idea to Research Paper” by Martha Larson @ngrams on Tuesday Oct, 11 at 16:35 - 17:30 in Pav. 4 R1.07 #ACMMultimedia2022
PhD School Talk @acmmm2022 announcement: “Multimodal Computing for a Healthy Society” by Ramesh Jain @jain49 on Tuesday Oct, 11 at 10:50 - 11:45 in Pav. 4 R1.07. #ACMMultimedia2022
Svebor Karaman retweeted
Great workshop in the AI for Social Good space organized by my colleagues at @Dataminr and @AyanMukhrjee at @VanderbiltU! The submission deadline is April 21 and it will be co-located with
@icwsm the first week of June.
My colleagues @tinybigdata, @AyanMukhrjee and I invite all to submit for the first Workshop on Social Media for Emergency Response, co-located with ICWSM (@icwsm).
Deadline: April 21st, 2022. See CFP for details.
2022sommer.github.io
Svebor Karaman retweeted
VizPol, developed by researchers at Columbia University, uses machine learning to identify the political symbols of the far-right or left econ.trib.al/z6utivp
Svebor Karaman retweeted
@ninaberman, @SusanEMcG, @SveborK, @ghoshbhaskar10, other amazing researchers at Columbia's DVMM lab, and I built an app called VizPol that helps reporters on the ground identify political symbols at demonstrations in real-time
cjr.org/tow_center/the-tow-c…
Svebor Karaman retweeted
An experience @ninaberman had at a political rally last year lead to an interdisciplinary research project at Columbia about tracking political symbols:
cjr.org/tow_center/political…
Big thanks to @samthielman for helping us get the blog post out!
Our (cc Federico Bartoli and @lisanti_g) paper "Scene-dependent Proposals for Efficient Person Detection" was just published in Pattern Recognition, get free access here: authors.elsevier.com/a/1Xx5D…
Svebor Karaman retweeted
Comprehensive coverage of topics including multimedia content understanding, multimodal human-human and human-computer interaction, multimedia social media, and multimedia system architecture and deployment. bit.ly/2F5Ge0k
The FCC is days away from voting to kill NetNeutrality, but Congress can still stop them. On December 12th we'll #BreakTheInternet to #SAVENETNEUTRALITY and #STOPTHEFCC: battleforthenet.com/breakthe…
We have released the code of our state-of-the-art approach on #reidentification on #github: github.com/glisanti/MCK-CCA. w/ @iBeppino @_iAc
Svebor Karaman retweeted
Check out our recent approach on person re-identification: goo.gl/9hlQ0L source code: goo.gl/SC4sB6 @_iAc @SveborK
Svebor Karaman retweeted
Check out our paper published in Multimedia Tools and Applications rdcu.be/oCTs with @SveborK and @pezzabros #train #safety
Svebor Karaman retweeted
Demo ready! Join us to try our real time wearable system for an improved museum experience! #acmmm16
Svebor Karaman retweeted
Col sistema del Progetto Mnemosine al Bargello è possibile tracciare l'attività dei visitatori per dare loro suggerimenti. #museiemotivi
Svebor Karaman retweeted
Useful datasets and classifiers released by the group of Prof. Shih-fu Chang in ACM MM'15: eventnet.ee.columbia.edu + mvso.cs.columbia.edu
Svebor Karaman retweeted
3rd Workshop on Web-scale Vision and Social Media @ ICCV 2015 [sites.google.com/site/vsm201…]. Full paper (6-8 p) D:25-09. Ext abs (2 p) D:15-10
Svebor Karaman retweeted
Everything you wanted to know about image tagging refinement & retrieval: "Socializing the Semantic Gap: A Survey..." goo.gl/FyOmoL















