Previous | 
Home
Section: Software
Software
When applicable, we provide the IDDN is the official number, which is obtained when registering the software at the APP (Agence de Protection des Programmes).
New Software
Aabot
Participant : Jonathan Delhumeau.
Aabot is a tool to facilitate annotation of large video databases. It's primary design focus has been for the annotation on commercials in two 6-month long TV databases. The software keeps a database of already annotated commercials and suggests when it finds a new probable instance. It also validates user annotations by suggesting similar existing commercials if it finds any which are similar by name or content. The user can then confirm the creation of new commercials or accept the correction if he was mistaken.
Aabot is accessed via a web-browser. It is mostly used by uploading and downloading an annotation file. An interactive HTML5 interface is also available when some user feedback is needed (during validation). It uses Peyote as an description / indexation engine.
First APP deposit: IDDN.FR.001.4200010.000.S.P.2012.000.20900 .
Peyote
Participants : Sébastien Campion, Jonathan Delhumeau [correspondent] , Hervé Jégou.
Peyote is a framework for Video and Image description, indexation and nearest neighbor search. It can be used as-is by a video-search or image-search front-end with the implemented descriptors and search modules. It can also be used via scripting for large-scale experimentation. Finally, thanks to its modularity, it can be used for scientific experimentation on new descriptors or indexation methods. Peyote is used in the Aabot software and was used for the Mediaeval Placing task [68] and the Trecvid Instance Search task.
First APP deposit: IDDN.FR.001.4200008.000.S.P.2012.000.20900 .
Watermarking Effective Key Length Evaluation
Participant : Teddy Furon [correspondent] .
This software was developed in collaboration with Patrick BAS (CNRS, Ecole Centrale de Lille)
Weckle is a software suite in Matlab and R for the numerical evaluation of the effective key length of watermarking schemes based on Spread Spectrum, a concept which was proposed in [22] , [23] .
Most active software started before 2012
Babaz
Participants : Jonathan Delhumeau, Guillaume Gravier, Hervé Jégou [correspondent] .
Babaz (http://babaz.gforge.inria.fr/ ) is a audio database management system with an audio-based search function, which is intended for audio-based search in video archives. First APP deposit: IDDN.FR.001.010006.000.S.P.2012.000.10000 . It is licensed under the terms of the GNU General Public License v3.0.
Bigimbaz
Participant : Hervé Jégou [correspondent] .
Bigimbaz is a platform originally developed in the Lear project-team, and now co-maintained by TexMex . It integrates several contributions on image description and large-scale indexing: detectors, descriptors, retrieval using bag-of-words and inverted files, and geometric verification.
BonzaiBoost
Participant : Christian Raymond [correspondent] .
The software homepage is available at http://bonzaiboost.gforge.inria.fr/ .
BonzaiBoost stands for boosting over small decisions trees. BonzaiBoost is a general purpose machine-learning program based on decision tree and boosting for building a classifier from text and/or attribute-value data. Currently one configuration of BonzaiBoost is ranked first on http://mlcomp.org a website which propose to compare several classification algorithms on many different datasets
Irisa_Ne
Participant : Christian Raymond [correspondent] .
Irisa_Ne is a couple of Named Entity tagger, one of them is based on CRF and the other HMM. It is dedicated to automatic transcriptions of speech. It does not take into account uppercase or punctuation and has no concept of sentences. However, they also manage texts with punctuation and capitalization.
Nero
Participant : Sébastien Campion [correspondent] .
The service is available at https://nero.irisa.fr .
Nero is an online Named Entities Recognition system. It is implemented within a web service that allows other member of the community to evaluate our results online without any client side setup. An HTTP Rest API, Shell and Python client are provided. The protocol used is HTTPS to secure the transactions between the user and the server. A user account is needed, which allow a fine monitoring. Usage are also limited to 100 thousand characters per account.
SURVA
Participants : Sébastien Campion [correspondent] , Jonathan Delhumeau.
Speed Up Robust Video Alignement enables to quickly and efficiently synchronize the same video with two coding and quality formats (i.e. without the same number of frame). First APP deposit: IDDN.FR.001.420009.000.S.P.2012.000.20900 .
PimPy
Participant : Sébastien Campion [correspondent] .
PimPy provides a convenient and high level API to manage common multimedia indexing tasks. It includes severals features. It is used, in particular
to retrieve video features, such as histogram, binarized DCT descriptor, SIFT, SURF, etc ;
for raw frame extraction, or video segment extraction and re-encoding ;
to search a video segment in another video (content based retrieval) ;
First APP deposit: IDDN.FR.001.260038.000.S.P.2011.000.40000
Pqcodes
Participant : Hervé Jégou [correspondent] .
This software is jointly maintained by Matthijs Douze, from Inria Grenoble.
Pqcodes is a library which implements the approximate k nearest neighbor search method of [83] based on product quantization. This software has been transferred to two companies (in August 2011 and May 2012, respectively).
The current version registered at the APP is IDDN.FR.001.220012.001.S.P.2010.000.10000 .
Yael
Participant : Hervé Jégou [correspondent] .
This software is jointly maintained by Matthijs Douze, from Inria Grenoble.
Yael is a C/python/Matlab library providing (multi-threaded, Blas/Lapack, low level optimization) implementations of computationally demanding functions. In particular, it provides very optimized functions for k-means clustering and exact nearest neighbor search. The library has been downloaded about 1000 times in 2012.
The current version registered at APP is IDDN.FR.001.220014.001.S.P.2010.000.10000.
IRISA News Topic Segmenter (irints)
Participants : Guillaume Gravier [correspondent] , Camille Guinaudeau, Pascale Sébillot, Anca-Roxana Simon.
This software is dedicated to unsupervised topic segmentation of texts and transcripts. The software implements several of our research methods and is particularly adapted for automatic transcripts. It provides topic segmentation capabilities virtually for any word-based language, with presets for French, English and German. The software has been licensed to several of our industrial partners.
Other softwares
Bag-of-colors , implements a technique to describe the images based on color.
Asares is a symbolic machine learning system that automatically infers, from descriptions of pairs of linguistic elements found in a corpus in which the components are linked by a given semantic relation, corpus-specific morpho-syntactic and semantic patterns that convey the target relation. IDDN.FR.001.0032.000.S.C.2005.000.20900 .
AnaMorpho detects morphological relations between words in many languages IDDN.FR.001.050022.000.S.P.2008.000.20900 .
DiVATex is a audio/video frame server. IDDN.FR.001.320006.000.S.P.2006.000.40000 ,
NaviTex is a video annotation tool. IDDN.FR.001.190034.000.S.P.2007.000.40000 ,
Telemex is a web service that enables TV and radio stream recording.
VidSig computes a small and robust video signature (64 bits per image).
VidSeg computes segmentation features such as cuts, dissolves, silences in audio track, changes of ratio aspect, monochrome images. IDDN.FR.001.250009.000.S.P.2009.000.40000 ,
Isec , web application used as graphical interface for image searching engines based on retrieval by content.
GPU-KMeans , implementation of k-means algorithm on graphical process unit (graphic cards)
Correspondence Analysis computes a factorial correspondence analysis (FCA) for image retrieval.
GPU Correspondence Analysis is an implementation of the previous software Correspondence Analysis on graphical processing unit (graphical card).
CAVIZ is an interactive graphical tool that allows to display and to extract knowledge from the results of a Correspondence Analysis on images.
Kiwi (standing for Keywords Extractor) is mostly dedicated to indexing and keyword extraction purposes.
Topic Segmenter , is a software dedicated to topic segmentation of texts and (automatic) transcripts.
S2E (Structuring Events Extractor) is a module which allows the automatic discovery of audiovisual structuring events in videos.
2pac builds classes of words of similar meanings (“semantic classes“) specific to the use that is made of them in that given topic. IDDN.FR.001.470028.000.S.P.2006.000.40000
Faestos (Fully Automatic Extraction of Sets of keywords for TOpic characterization and Spotting) is a tool composed of a sequence of statistical treatments that extracts from a morpho-syntactically tagged corpus sets of keywords that characterize the main topics that corpus deals with. IDDN.FR.001.470029.000.S.P.2006.000.40000
Fishnet is an automatic web pages grabber associated with a specific theme.
Match Maker , semantic relation extraction by statistical methods.
python-geohash is an implementation of the Geometric Hashing algorithm of [90] to check if geometrical consistency between pairs of images.
AVSST is an Automatic Video Stream Structuring Tool. First, it allows the detection of repetitions in a TV stream. Second, a machine learning method allows the classification of programs and inter-programs such as advertisements, trailers, etc. Finally, the electronic program guide is synchronized with the right timestamps based on dynamic time warping. A graphical user interface is provided to manage the complete workflow.
TVSearch is a content based retrieval search engine used to search and propagate manual annotation such as advertisement in a TV corpora.
Samusa detects speech and/or musical segment in multimedia content.
kertrack is a visual graphical interface for tracking visual targets based on particle filter tracking or based on mean-shift.
mozaic2d creates of spatio-temporal mosaic based on dominant motion compensation.