1,720,974 research outputs found
Towards automatic acquisition of high-level 3D models from images
No full textNei tempi recenti abbiamo assistito a un crescente interesse nella modellazione automatica da immagine. Mentre gli studi recenti nell'ambito della ricostruzione tridimensionale si sono concentrata soprattutto sull'estrazione di rappresentazioni dense e accurate di oggetti catturati tramite foto o video, il sostenuto interesse verso software di modellazione accessibile è una forte riprova del grande bisogno di rappresentazioni astratte e compatte degli oggetti.
In questa tesi, il problema dell'estrazione di modelli di alto livello a partire dalle immagini viene discusso in dettaglio.
Nella prima parte viene introdotta una pipeline di "Structure from Motion". A partire dai risultati di tale pipeline, vengono studiati due differenti approcci per la generazione di modelli di alto livello.
Nel primo approccio viene innanzitutto introdotto un nuovo algoritmo di stereo multivista per produrre una nuvola di punti densa e accurata. Successivamente viene presentato un sistema di ricerca e reperimento di mesh, basato su segmentazione e un algoritmo di tipo "Bag of Words".
Nel secondo approccio, la nuvola di punti sparsa proveniente dalla pipeline di "Structure from Motion" viene
descritta da piani e aree planari convesse. Le aree planari sono una rappresentazione compatta e intermedia della scena. Entrambe le parti della tesi mirano ad assottigliare il divario tra acquisizione e interpretazione di una scena, attraverso la definizione di rappresentazioni ad alto livello ottenute tramite strategie molto diverse tra loro.In recent years there has been a surge of interest in automatic modeling from images. While the current state of the art in three-dimensional reconstruction has
focused on the recovery of dense and accurate representations of objects imaged through pictures or video, the sustained interest in accessible modeling software is a strong evidence of an untapped general need for compact, abstract representations
of objects. In this thesis, the problem of producing high level models starting from images is discussed in details. In the first part, an automatic uncalibrated Structure from Motion pipeline is presented. Starting from the output of the pipeline, two different approaches of generating high-level renditions are studied. The first approach employs a novel Multiple view Stereo algorithm to produce a dense and accurate point cloud. A retrieval system for meshes, based on segmentation and Bag of Words, is then introduced. In the latter approach, the sparse Structure from Motion point cloud is fitted by planes and planar patches. Planar patches
are a compact, intermediate representation of the scene. Both branches of the thesis aim to narrow the gap between scene acquisition and interpretation, through
the definition of high level renditions produced by very different strategies
Photo-consistent planar patches from unstructured cloud of points
No full textPlanar patches are a very compact and stable intermediate representation of 3D scenes, as they are a good starting point for a complete automatic reconstruction of surfaces. This paper presents a novel method for extracting planar patches from an unstructured cloud of points that is produced by a typical structure and motion pipeline. The method integrates several constraints inside J-linkage, a robust algorithm for multiple models fitting. It makes use of information coming both from the 3D structure and the images. Several results show the effectiveness of the proposed approach
Robust multiple structures estimation with J-linkage
No full textThis paper tackles the problem of fitting multiple instances of a model to data corrupted by noise and outliers. The proposed solution is based on random sampling and conceptual data representation. Each point is represented with the characteristic function of the set of random models that fit the point. A tailored agglomerative clustering, called J-linkage, is used to group points belonging to the same model. The method does not require prior specification of the number of models, nor it necessitate parameters tuning. Experimental results demonstrate the superior performances of the algorithm
Hierarchical structure-and-motion recovery from uncalibrated images
No full textThis paper addresses the structure-and-motion problem, that requires to find camera motion and 3D structure from point matches. A new pipeline, dubbed Samantha, is presented, that departs from the prevailing sequential paradigm and embraces instead a hierarchical approach. This method has several advantages, like a provably lower computational complexity, which is necessary to achieve true scalability, and better error containment, leading to more stability and less drift. Moreover, a practical autocalibration procedure allows to process images without ancillary information. Experiments with real data assess the accuracy and the computational efficiency of the method
Visual vocabulary signature for 3D object retrieval and partial matching
No full textIn this paper a novel object signature is proposed for 3D object retrieval and partial matching. A part-based representation is obtained by partitioning the objects into subparts and by characterizing each segment with different geometric descriptors. Therefore, a Bag ofWords framework is introduced by clustering properly such descriptors in order to define the so called 3D visual vocabulary. In this fashion, the object signature is defined as a histogram of 3D visual word occurrences. Several examples on the Aim@Shape watertight dataset demonstrate the versatility of the proposed method in matching either 3D objects with articulated shape changes or partially occluded or compound objects. In particular, a comparison with the methods that participated to the Shape Retrieval contest 2007 (SHREC) reports satisfactory results for both object retrieval and partial matching
A bag of words approach for 3D object categorization
No full textIn this paper we propose a novel framework for 3D object categorization. The object is modeled it in terms of its sub-parts as an histogram of 3D visual word occurrences. We introduce an effective method for hierarchical 3D object segmentation driven by the minima rule that combines spectral clustering - for the selection of seed-regions - with region growing based on fast marching. The front propagation is driven by local geometry features, namely the Shape Index. Finally, after the coding of each object according to the Bag-of-Words paradigm, a Support Vector Machine is learnt to classify different objects categories. Several examples on two different datasets are shown which evidence the effectiveness of the proposed framewor
A matrix decomposition perspective on calibrated photometric stereo
Get PDFLeveraging on recent advances in robust matrix decomposition, we revisit Lambertian photometric stereo as a robust low-rank matrix recovery problem with both missing and corrupted entries, tailoring Grasta and R-GoDec to normal surface estimation. A method to automatically detect shadows is proposed. The performance of different robust matrix completion techniques are analyzed on the challenging DiLiGenT datasets
- …
