On the use of simple geometric descriptors provided by RGB-D sensors for re-identification

[EN]The re-identification problem has been commonly accomplished using appearance features based on salient points and color information. In this paper, we focus on the possibilities that simple geometric features obtained from depth images captured with RGB-D cameras may offer for the task, particularly working under severe illumination conditions. The results achieved for different sets of simple geometric features extracted in a top-view setup seem to provide useful descriptors for the re-identification task, which can be integrated in an ambient intelligent environment as part of a sensor network.

Evaluation of LBP and HOG descriptors for clothing attribute description

[EN]In this work an experimental study about the capability of the LBP, HOG descriptors and color for clothing attribute classification is presented. Two different variants of the LBP descriptor are considered, the original LBP and the uniform LBP. Two classifiers, Linear SVM and Random Forest, have been included in the comparison because they have been frequently used in clothing attributes classification. The experiments are carried out with a public available dataset, the clothing attribute dataset, that has 26 attributes in total. The obtained accuracies are over 75% in most cases, reaching 80% for the necktie or sleeve length attributes.

Exploring the use of local descriptors for fish recognition in LifeCLEF 2015

[EN]This paper summarizes the proposal made by the SIANI team for the LifeCLEF 2015 Fish task. The approach makes use of standard detection techniques, applying a multiclass SVM based classifier on large enough Regions Of Interest (ROIs) automatically extracted from the provided video frames. The selection of the detection and classification modules is based on the best performance achieved for the validation dataset consisting of 20 annotated videos. For that dataset, the best classification achieved for an ideal detection module, reaches an accuracy around 40%.

Multi-scale score level fusion of local descriptors for gender classification in the wild

The 2015 FRVT gender classification (GC) report evidences the problems that current approaches tackle in situations with large variations in pose, illumination, background and facial expression. The report suggests that both commercial and research solutions are hardly able to reach an accuracy over 90% for The Images of Groups dataset, a proven scenario exhibiting unrestricted or in the wild conditions. In this paper, we focus on this challenging dataset, stepping forward in GC performance by observing: 1) recent literature results combining multiple local descriptors, and 2) the psychophysics evidences of the greater importance of the ocular and mouth areas to solve this task...

Temporal patterns of the seagrass "Cymodocea nodosa" and depth-related patterns of the green algae "Caulerpa prolifera" in the Canarian Archipelago

Máster Oficial en Gestión Costera

Effect of culture system and diet in early larval stages over skeletal deformities development in gilthead sea bream

[ES] Main deformities such as lordosis, opercular deformities and upper/lower jaws shortening are considered as quality descriptors in commercial marine fish fry production and seem to be related with larval culture conditions in early larval stages. The aim of this work was to obtain information about the contribution of the diet and rearing system to the apparition of these abnormalities in gilthead sea bream (Sparus aurata) larvae in semi-industrial scale facilities. For that purpose, two different larval rearing systems semi-intensive and intensive were compared; besides, two different rotifer enrichments, DHA Protein Selco, (Inve Aquaculture, Dendermonde, Belgium) (R1) and Red Pepper Paste, (Bernaqua bvba, Turnhout, Belgium) were tested in the intensive system. Biochemical composition of larvae, preys and commercial products was analysed. At 50 days post hatching six hundred fish per treatment were individually studied under stereoscope and deformity frequency recorded. Besides at 95 days post hatching fry were soft X ray monitored. Both rotifer enrichment and rearing system affected survival, growth and deformity frequency. Rearing system did not affect total larvae fatty acid content except at 20 dah, where DHA were significantly higher and EPA significantly lower in Semi-intensive system. A significantly lower percentage of deformity rates together with better survival and growth were obtained in the semi-intensive system. In dietary treatment, rotifer enrichment significantly affected larval survival. R1 rotifers enrichment significantly (P<0.05) improved survival when compared to fed R2 larvae. The content of DPA was significantly (P<0.05) higher in R2 fed larvae reflecting the R2 rotifers content of this fatty acid. The level of this FA tended to decrease in concordance with the rotifers replacement by artemia in the diet. The effects n-3-HUFA and DPA (22:5n-6) over larval survival and skeletal deformities development is discussed.

Effect of rearing techniques over, survival growth and skeletal abnormalities development in red porgy (Pagrus Pagrus) larvae

[EN] Red porgy is one of the most interested new species for Spanish and other Mediterranean countries. Although no industrial procedures for fry production has been yet developed. The aim of this work was to develop an industrial scale larval rearing protocols testing the viability of two different rearing techniques (semi-intensive vs intensive) in pilot scale facilities. The second objective was to obtain information about the contribution of rearing system to the apparition of morphological abnormalities such as lordosis, opercular deformities and upper/lower jaws shortening which are considered as quality descriptors in commercial marine fish fry production and seem to be related with larval culture conditions in early larval stages. For that purpose, two different larval rearing systems semi-intensive and intensive were compared using the same live feed enrichments. Biochemical composition of larvae, preys and commercial products was analysed. At 50 days post hatching six hundred fish per treatment was individually studied under stereoscope and abnormalities frequency recorded. At 95 days post hatching fry were soft X ray monitored as well. Survival and abnormalities frequency were similar between treatments although a better growth in terms of total length was obtained in the semi-intensive system.

Development of skeletal deformities in Gilthead sea bream (Sparus aurata) reared under different larval culture and dietary conditions

[EN] Main deformities such as lordosis, opercular deformities and upper/lower jaws shortening are considered as quality descriptors in commercial marine fish fry production and seem to be related at least with larval culture conditions in early larval stages. The aim of this work was to obtain information about the contribution of the diet and rearing system to the apparition of these abnormalities in gilthead sea bream (Sparus aurata) larvae in semi-industrial scale facilities. For that purpose, two different larval rearing systems semi-intensive and intensive were compared by duplicate and with the same live feed enrichments; besides, two different rotifer enrichments were tested in an intensive system. Biochemical composition of larvae, preys and commercial products was analysed. At 50 days post hatching six hundred fish per treatment were individually studied under stereoscope and abnormalities frequency recorded. At 95 days post hatching fry were soft X ray monitored as well. Survival and malformation frequency were significantly different between treatments, the effect of diet and system are discussed. A significantly lower percentage of deformity rates together with better survival and growth were obtained in the semi-intensive system, whereas the rotifer enrichment significantly affected larval survival.

Prototipo de herramienta para el inventariado automático de piezas de Lego basado en visión artificial

Permitida la difusión del código bajo los términos de la licencia BSD de tres cláusulas.

Improving gender classification accuracy in the wild

[EN]In this paper, we focus on gender recognition in challenging large scale scenarios. Firstly, we review the literature results achieved for the problem in large datasets, and select the currently hardest dataset: The Images of Groups. Secondly, we study the extraction of features from the face and its local context to improve the recognition accuracy. Diff erent descriptors, resolutions and classfii ers are studied, overcoming previous literature results, reaching an accuracy of 89.8%.

An study on re-identification in RGB-D imagery

[EN]Re-identi fication is commonly accomplished using appearance features based on salient points and color information. In this paper, we make an study on the use of di fferent features exclusively obtained from depth images captured with RGB-D cameras. The results achieved, using simple geometric features extracted in a top-view setup, seem to provide useful descriptors for the re-identi fication task.

People semantic description and re-identification from point cloud geometry

The automatic extraction of biometric descriptors of anonymous people is a challenging scenario in camera networks. This task is typically accomplished making use of visual information. Calibrated RGBD sensors make possible the extraction of point cloud information. We present a novel approach for people semantic description and re-identification using the individual point cloud information. The proposal combines the use of simple geometric features with point cloud features based on surface normals.

A study for the self similarity smile detection

[EN]This work makes an extensive experimental study of smile detection testing the Local Binary Patterns (LBP) combined with self similarity (LAC) as main descriptors of the image, along with the powerful Support Vector Machines classifier. Results show that error rates can be acceptable and the self similarity approach for the detection of smiles is suitable for real-time interaction, although there is still room for improvement.