Search

In this paper we aim to explore the potential of Deep Convolutional Neural Networks (DCNNs) on food image segmentation where semantic segmentation paradigm is used to separate food regions from the non-food regions. Specifically, we are interested in evaluating the performance of an efficient DCNN with respect to variability in illumination conditions that can be found in food images taken in real scenarios. To this end we have designed an experimental setup where the network is trained on images rendered as if they were taken under nine different illuminants. We evaluate the food vs. non-food segmentation performance of the network in terms of standard Intersection over Union (IoU) measure. The results of this experimentation are reported and discussed.

Lo scopo di questa tesi è di studiare due principali categorie di algoritmi per la detection di oggetti e il loro uso in particolari applicazioni. La prima categoria esaminata riguarda approcci basati su Keypoint. Diversi esperimenti comparativi vengono eseguiti all'interno della pipeline standard del modello di test MPEG CDVS e viene proposta una pipeline estesa che fa uso di informazione colore. La seconda categoria di object detectors oggetto di indagine si basa su Reti neurali convoluzionali. In particolare, vengono affrontate due applicazioni di reti neurali convoluzionali per il riconoscimento di oggetti. Il primo riguarda il riconoscimento di loghi commerciali. Due pipeline di classificazione sono progettate e testate su un set di immagini raccolte da Flickr. La prima architettura utilizza una rete pre-addestrata come feature extractor e raggiunge risultati comparabili a quelli di algoritmi basati Keypoint. La seconda architettura si avvale di una rete neurale che supera le performances di metodi stato dell'arte basati su Keypoint. L'altra applicazione esaminata è la categorizzazione di dipinti che consiste nell'associare l'autore, nell'assegnare un dipinto alla scuola o al movimento artistico a cui appartiene, e classificare il genere del dipinto, ad es. paesaggio, ritratto, illustrazione ecc. Per affrontare questo problema, viene proposta una struttura di rete neurale multibranch e multitask che beneficia dell'uso congiunto di approcci basati su keypoint e di features neurali. In entrambe le applicazioni viene anche esaminato l'uso di tecniche di data augmentation per ampliare il training set. In particolare per i dipinti, un algoritmo di trasferimento di stile pittorico basato su reti neurali viene sfruttato per generare quadri sintetici da utilizzare in fase di training., The aim of this thesis is to study two main categories of algorithms for object detection and their use in particular applications. The first category that is investigated concerns Keypoint-based approaches. Several comparative experiments are performed within the standard testing pipeline of the MPEG CDVS Test Model and an extended pipeline which make use of color information is proposed. The second category of object detectors that is investigated is based on Convolutional Neural Networks. Two applications of Convolutional Neural Networks for object recognition are in particular addressed. The first concerns logo recognition. Two classification pipelines are designed and tested on a real-world dataset of images collected from Flickr. The first architecture makes use of a pre-trained network as feature extractor and it achieves comparable results keypoint based approaches. The second architecture makes use of a tiny end-to-end trained Neural Network that outperformed state-of-the-art keypoint based methods. The other application addressed is Painting Categorization. It consists in associating the author, assigning a painting to the school or art movement it belongs to, and categorizing the genre of the painting, e.g. landscape, portrait, illustration etc. To tackle this problem, a novel multibranch and multitask Neural Network structure is proposed which benefit from joint use of keypoint-based approaches and neural features. In both applications the use of data augmentation techniques to enlarge the training set is also investigated. In particular for paintings, a neural style transfer algorithm is exploited for generating synthetic paintings to be used in training.

We propose a novel modular CNN architecture that provides semantic segmentation and understanding of outdoor street environment images. Our solution processes a 512x1024 resolution image on a single Titan Xp GPU at 37.4 FPS attaining 70.4% IoU on the Cityscapes test dataset.

In this paper we propose a method for logo recognition using deep learning. Our recognition pipeline is composed of a logo region proposal followed by a Convolutional Neural Network (CNN) specifically trained for logo classification, even if they are not precisely localized. Experiments are carried out on the FlickrLogos-32 database, and we evaluate the effect on recognition performance of synthetic versus real data augmentation, and image pre-processing. Moreover, we systematically investigate the benefits of different training choices such as class-balancing, sample-weighting and explicit modeling the background class (i.e. no-logo regions). Experimental results confirm the feasibility of the proposed method, that outperforms the methods in the state of the art.

In this paper we propose a method for logo recognition based on Convolutional Neural Networks, instead of the commonly used keypoint-based approaches. The method involves the selection of candidate subwindows using an unsupervised segmentation algorithm, and the SVM-based classification of such candidate regions using features computed by a CNN. For training the neural network we augment the training set with artificial transformations, while for classification we exploit a query expansion strategy to increase the recall rate. Experiments were performed on a publicly-available dataset that was also corrupted in order to investigate the robustness of the proposed method with respect to blur, noise and lossy compression.

Local Visual detectors and descriptors have been studied for many years, but their applications (e.g. mobile visual search) in large volume, low-cost, low-power embedded systems have been limited or negligible to date. One reason is the lack of a worldwide industry standard. MPEG Compact Descriptors for Visual Search (CDVS) working group filled this gap by defining a high-performance extraction stage and the bitstream syntax at its output in order to achieve interoperability between different implementations of clients and servers. In a previous work, we presented an analysis of various gray-level interest point detection and description algorithms, which was also contributed to CDVS. This work extends the previous analysis using the MPEG CDVS Test Model framework to consider additional detectors and the use of color descriptors.

Local Visual detectors and descriptors have been studied for many years, but their applications, for example on visual search, for large volume, low cost, embedded systems have been limited or negligible to date. One reason is the lack of an industry standard applicable at worldwide level. MPEG Compact Descriptors for Visual Search (CDVS) working group is filling the gap by defining a highly performing extraction stage and the bitstream syntax at its output. As a contribution to such process, this work reviews existing methods for interest point detection and description. They are compared on several databases, exploiting the pairwise matching test environment devised by the MPEG CDVS standardization group. © 2013 IEEE.