Johann Poignant

Learn More
We present in this article a video OCR system that detects and recognizes overlaid texts in video as well as its application to person identification in video documents. We proceed in several steps. First, text detection and temporal tracking are performed. After adaptation of images to a standard OCR system, a final post-processing combines multiple(More)
We propose an approach for unsupervised speaker identification in TV broadcast videos, by combining acoustic speaker diarization with person names obtained via video OCR from overlaid texts. Three methods for the propagation of the overlaid names to the speech turns are compared, taking into account the co-occurence duration between the speaker clusters and(More)
Experiment 1 recorded the effects of single (doses of 1, 5, 10, and 20 mg/kg) and repeated intraperitoneal injections (10 mg/kg) of amineptine (a tricyclic antidepressant drug) on the performance of albino rats in differential reinforcement of low rate (DRL) of 30 s, fixed-interval (FI) of 60 s, and signalled continuous reinforcement (CRF-SD) schedules. In(More)
Most state-of-the-art approaches address speaker diarization as a hierarchical agglomerative clustering problem in the audio domain. In this paper, we propose to revisit one of them: speech turns clustering based on the Bayesian Information Criterion (a.k.a. BIC clustering). First, we show how to model it as an integer linear programming (ILP) problem. Its(More)
The Repere challenge is a project aiming at the evaluation of systems for supervised and unsupervised multimodal recognition of people in TV broadcast. In this paper, we describe, evaluate and discuss QCompere consortium submissions to the 2012 Repere evaluation campaign dry-run. Speaker identification (and face recognition) can be greatly improved when(More)
Identifying speakers in TV broadcast in an unsupervised way (i.e., without biometric models) is a solution for avoiding costly annotations. Existing methods usually use pronounced names, as a source of names, for identifying speech clusters provided by a diarization step but this source is too imprecise for having sufficient confidence. To overcome this(More)
Persons’ identification in TV broadcast is one of the main tools to index this type of videos. The classical way is to use biometric face and speaker models, but, to cover a decent number of persons, costly annotations are needed. Over the recent years, several works have proposed to use other sources of names for identifying people, such as pronounced(More)
We studied the biochemical and pharmacologic modes of action of piribedil and apomorphine in the rat. Although both drugs have many points in common, they are also different in many of their manifestations. Apomorphine causes high-intensity, short-duration stereotyped behavior; it is distributed within the brain in uneven fashion, the striatum being the(More)
Structure-activity relationships in the classical antidepressant (imipramine-like) series show a relative lack of specificities: Compounds should simply have a nucleus consisting of two phenyl rings and a third, seven-member central ring. This central ring may have one, several, or no heteroatoms, and it may or may not be saturated. The side chain may be(More)