Tag Archives: MICC

MNEMOSYNE Project

MNEMOSYNE is a three years research project co-funded by the MICC – University of Florence and the Tuscany – European Social Fund. The project is about the study and experimentation of smart environments for the protection and promotion of artistic and cultural heritage. It adopts natural interaction paradigms for access and manipulation of multimedia information and rely on the passive analysis of visitors behaviors [1] to estimate their interests.

mnemosyneThe current vision techniques applied in cultural heritage environments, such as museums, usually provide solutions for protection by detecting situations of potential risk and then notify operators responsible for safety. The idea of the project is to use techniques derived from video-surveillance scenarios to design an automatic profiling system capable of understanding the personal interest of many visitors.

The computer vision system monitors and analyzes the movements and behaviors of visitors in the museum (through the use of fixed cameras) in order to extract a profile of interests for each visitors. This profile of interest is then used to personalize the delivery of in-depth multimedia content enabling an augmented museum experience. Visitors interact with the multimedia content through a large interactive table installed inside the museum. The project also includes the integration of mobile devices (such as smartphones or tablets) offering a take-away summary of the visitor experience and suggesting possible theme-related paths in the collection of the museum or in other places of the city.

My work in this project was to build the back-end and computer vision systems, especially the passive profiling of the visitors inside the museum, with related research themes such as Re-Identification [2, 3, 4], Tracking [5], and Person Detection [6]. The whole MNEMOSYNE system [7] is currently under deployment at the Bargello Museum in Florence.

Related publications

[1] [pdf] [doi] S. Karaman, A. D. Bagdanov, G. D’Amico, L. Landucci, A. Ferracani, D. Pezzatini, and A. Del Bimbo, “Passive Profiling and Natural Interaction Metaphors for Personalized Multimedia Museum Experiences,” in MM4CH’13 – New Trends in Image Analysis and Processing – ICIAP 2013, Naples, Italy: Springer, 2013, p. 247–256.
[Bibtex]
@incollection{karaman2013passive,
title = {Passive Profiling and Natural Interaction Metaphors for Personalized Multimedia Museum Experiences},
author = {Karaman, Svebor and Bagdanov, Andrew D and D’Amico, Gianpaolo and Landucci, Lea and Ferracani, Andrea and Pezzatini, Daniele and Del Bimbo, Alberto},
booktitle = {MM4CH'13 - New Trends in Image Analysis and Processing -- ICIAP 2013},
doi = {10.1007/978-3-642-41190-8_27},
pages = {247--256},
address = {Naples, Italy},
year = {2013},
note={Oral Presentation},
publisher = {Springer}
}
[2] [pdf] [doi] S. Karaman and A. D. Bagdanov, “Identity Inference: Generalizing Person Re-identification Scenarios,” in Computer Vision – ECCV 2012. Workshops and Demonstrations, A. Fusiello, V. Murino, and R. Cucchiara, Eds., Firenze, Italy: Springer Berlin Heidelberg, 2012, vol. 7583, pp. 443-452.
[Bibtex]
@incollection{karamanIdInf2012,
isbn={978-3-642-33862-5},
booktitle={Computer Vision – ECCV 2012. Workshops and Demonstrations},
volume={7583},
series={Lecture Notes in Computer Science},
editor={Fusiello, Andrea and Murino, Vittorio and Cucchiara, Rita},
doi={10.1007/978-3-642-33863-2_44},
title={Identity Inference: Generalizing Person Re-identification Scenarios},
url={http://dx.doi.org/10.1007/978-3-642-33863-2_44},
publisher={Springer Berlin Heidelberg},
author={Karaman, Svebor and Bagdanov, Andrew D.},
pages={443-452},
address = {Firenze, Italy},
note={Oral Presentation. Best Paper Award},
year={2012}
}
[3] [pdf] [doi] S. Karaman, G. Lisanti, A. D. Bagdanov, and A. Del Bimbo, “From Re-identification to Identity Inference: Labeling Consistency by Local Similarity Constraints,” in Person Re-Identification, S. Gong, M. Cristani, S. Yan, and C. C. Loy, Eds., Springer London, 2014, pp. 287-307.
[Bibtex]
@incollection{KaramanReID2014,
author = {Karaman, Svebor and Lisanti, Giuseppe and Bagdanov, Andrew D. and Del Bimbo, Alberto},
title = {From Re-identification to Identity Inference: Labeling Consistency by Local Similarity Constraints},
booktitle = {Person Re-Identification},
series = {Advances in Computer Vision and Pattern Recognition},
editor = {Gong, Shaogang and Cristani, Marco and Yan, Shuicheng and Loy, Chen Change},
isbn = {978-1-4471-6295-7},
doi = {10.1007/978-1-4471-6296-4_14},
url = {http://dx.doi.org/10.1007/978-1-4471-6296-4_14},
publisher = {Springer London},
keywords = {Re-identification; Identity inference; Conditional random fields; Video surveillance},
pages = {287-307},
language = {English},
year = {2014}
}
[4] [pdf] [doi] S. Karaman, G. Lisanti, A. D. Bagdanov, and A. D. Bimbo, “Leveraging local neighborhood topology for large scale person re-identification,” Pattern Recognition, vol. 47, iss. 12, pp. 3767-3778, 2014.
[Bibtex]
@article{karaman2014leveraging,
title = "Leveraging local neighborhood topology for large scale person re-identification ",
journal = "Pattern Recognition ",
volume = "47",
number = "12",
pages = "3767 - 3778",
year = "2014",
note = "",
issn = "0031-3203",
doi = "10.1016/j.patcog.2014.06.003",
url = "http://www.sciencedirect.com/science/article/pii/S0031320314002258",
author = "Svebor Karaman and Giuseppe Lisanti and Andrew D. Bagdanov and Alberto Del Bimbo",
keywords = "Re-Identification",
keywords = "Conditional Random Field",
keywords = "Semi-supervised",
keywords = "\{ETHZ\}",
keywords = "\{CAVIAR\}",
keywords = "3DPeS",
keywords = "\{CMV100\} "
}
[5] [pdf] [doi] A. D. Bagdanov, A. Del Bimbo, D. Di Fina, S. Karaman, G. Lisanti, and I. Masi, “Multi-Target Data Association using Sparse Reconstruction,” in Proc. of International Conference on Image Analysis and Processing (ICIAP), Naples, Italy, 2013, pp. 239-248.
[Bibtex]
@inproceedings{DBLMKD13,
author = {Bagdanov, Andrew D. and Del Bimbo, Alberto and Di Fina, Dario and Karaman, Svebor and Lisanti, Giuseppe and Masi, Iacopo},
title = {Multi-Target Data Association using Sparse Reconstruction},
booktitle = {Proc. of International Conference on Image Analysis and Processing (ICIAP)},
year = {2013},
address = {Naples, Italy},
pages = {239-248},
note={Poster},
doi = {10.1007/978-3-642-41184-7_25},
publisher = {Springer Berlin Heidelberg},
keywords = {Data association; multi-target tracking; sparse methods; video surveillance},
url = {http://www.micc.unifi.it/publications/2013/DBLMKD13}
}
[6] [pdf] F. Bartoli, G. Lisanti, S. Karaman, A. D. Bagdanov, and A. Del Bimbo, “Unsupervised scene adaptation for faster multi-scale pedestrian detection,” in 22nd International Conference on Pattern Recognition (ICPR), Stockholm, Sweden, 2014.
[Bibtex]
@InProceedings{bartoliicpr2014,
author = {Bartoli, Federico and Lisanti, Giuseppe and Karaman, Svebor and Bagdanov, Andrew D. and Del Bimbo, Alberto},
title = {Unsupervised scene adaptation for faster multi-scale pedestrian detection},
note = {Oral presentation},
booktitle = {22nd International Conference on Pattern Recognition (ICPR)},
address = {Stockholm, Sweden},
year = {2014}
}
[7] [pdf] [doi] S. Karaman, A. Bagdanov, L. Landucci, G. D’Amico, A. Ferracani, D. Pezzatini, and A. Del Bimbo, “Personalized multimedia content delivery on an interactive table by passive observation of museum visitors,” Multimedia Tools and Applications, pp. 1-25, 2014.
[Bibtex]
@article{karaman2014mtap,
year={2014},
issn={1380-7501},
journal={Multimedia Tools and Applications},
doi={10.1007/s11042-014-2192-y},
title={Personalized multimedia content delivery on an interactive table by passive observation of museum visitors},
url={http://dx.doi.org/10.1007/s11042-014-2192-y},
publisher={Springer US},
keywords={Computer vision; Video surveillance; Cultural heritage; Multimedia museum; Personalization; Natural interaction; Passive profiling},
author={Karaman, Svebor and Bagdanov, AndrewD. and Landucci, Lea and D’Amico, Gianpaolo and Ferracani, Andrea and Pezzatini, Daniele and Del Bimbo, Alberto},
pages={1-25},
language={English}
}

Supervised graduate students

I have supervised several graduate students (Laëtitia Letoupin, Antonin Molières, Alexis Malécot, Ho Tien Lam) for the development of a Video Indexing tool in C++/Qt while doing my PhD in Bordeaux.

Since joining the MICC, I have supervised multiple italian students on different projects:

  • Alessio Benevieri, Railway Wagon Id Automatic recognition
  • Federico Bartoli, Master thesis “Fast pedestrian detection via geometric and Soft Cascade approximation” [1]

  • Claudio Tortorici, Master thesis “Relaxed Decision Trees over multiple Taxonomies for Visual Recognition”

  • Giovanni Giunto, Master thesis “Towards Spatial Codebook-free Methods for Image Classification”

  • Andrea Ciolini, Master thesis “Object Detection on Low Power Devices” [2]

[1] [pdf] F. Bartoli, G. Lisanti, S. Karaman, A. D. Bagdanov, and A. Del Bimbo, “Unsupervised scene adaptation for faster multi-scale pedestrian detection,” in 22nd International Conference on Pattern Recognition (ICPR), Stockholm, Sweden, 2014.
[Bibtex]
@InProceedings{bartoliicpr2014,
author = {Bartoli, Federico and Lisanti, Giuseppe and Karaman, Svebor and Bagdanov, Andrew D. and Del Bimbo, Alberto},
title = {Unsupervised scene adaptation for faster multi-scale pedestrian detection},
note = {Oral presentation},
booktitle = {22nd International Conference on Pattern Recognition (ICPR)},
address = {Stockholm, Sweden},
year = {2014}
}
[2] [pdf] A. Ciolini, L. Seidenari, S. Karaman, and A. Del Bimbo, “Efficient Hough Forest Object Detection for Low-power Devices,” in IEEE First International Workshop on Wearable and Ego-vision Systems for Augmented Experience (WEsAX), 2015.
[Bibtex]
@inproceedings{ciolini2015,
author = {Ciolini, Andrea and Seidenari, Lorenzo and Karaman, Svebor and Del Bimbo, Alberto},
title = {Efficient Hough Forest Object Detection for Low-power Devices},
booktitle = {IEEE First International Workshop on Wearable and Ego-vision Systems for Augmented Experience (WEsAX)},
year = {2015}
}

About me

I am a French Computer Vision and Machine Learning researcher, currently a  Research Manager at Dataminr. Previously, I spent three years as a PostDoc at the MICC (Media Integration and Communication Center) of the University of Florence in Italy, and five years as an Associate Research Scientist in the DVMM Lab at Columbia University.

Research themes

My research themes are image and video analysis, computer vision, and machine learning. I am particularly interested in semantic concept recognition in images and videos.

I did my Ph.D. at the LaBRI – University of Bordeaux, under the supervision of Jenny Benois-Pineau and Rémi Mégret. During my Ph.D. thesis, I worked on human activity recognition by Hidden Markov Models (HMM) in videos recorded from a wearable device within the IMMED project. I have also developed an object recognition approach in the Bag-of-Visual-Words framework which integrates spatial information within semi-local features: the Graph-Words. I defended my Ph.D. entitled “Indexing of Activities in Wearable Videos: Application to Epidemiological Studies of Aged Dementia” in 2011.

While at the MICC, I have been highly involved in the MNEMOSYNE project. In this project, multiple aspects of computer vision such as person detection, person tracking, and re-identification are used to passively profile the interests of visitors in a museum to provide personalized multimedia content delivery. I was also working on more general image and video classification problems.

At the DVMM Lab, I have been working mostly on large-scale image indexing and retrieval problems but I also published works on other projects such as social media understanding, grounding, scene graph generation, visual parsing, and GAN detections…

At Dataminr, I’m working on computer vision and multimodal-related problems.

Keywords

Computer Vision, Machine Learning, Image Analysis, Video Analysis, Video Indexing, Object Recognition, Person Detection, Re-Identification, Passive Profiling, Behavior Analysis, Action Recognition…