Loading...
7 results
Search Results
Now showing 1 - 7 of 7
- Improving Point Cloud to Surface Reconstruction with Generalized Tikhonov RegularizationPublication . Guarda, André; Bioucas-Dias, José M.; M. M. Rodrigues, Nuno; Pereira, Fernando; Bernardo Pereira, Fernando ManuelPoint cloud rendering has a vital role in the user Quality of Experience for applications adopting point cloud based representations. While this is not a new area, it has recently become more relevant with the recent interest on point cloud coding by major standardization groups, notably JPEG and MPEG. The screened Poisson surface reconstruction is a state-ofthe- art technique for generating a watertight surface mesh from the point cloud samples. While its screening component allows the surface to better fit the cloud points, this fitting may lead to undesired artifacts in the surface, notably when the point cloud is noisy. This paper proposes to improve this reconstruction method by making it more robust to noise by adopting a generalized Tikhonov regularization term. The proposed regularization approach smooths regions that should be flat while keeping the important details in the edges, thus creating more pleasant surface reconstructions.
- Studying the Benefits of a New JPEG AI Profile for the JPEG PCC Verification Model: ISO/IEC JTC 1/SC29/WG1 M100115Publication . Seleem, Abdelrahman; Guarda, André; Rodrigues, Nuno; Pereira, FernandoContext and Objective: The current JPEG PCC VM color coding approach first projects the PC color onto 2D images, then uses JPEG AI to code the 2D images.
- IT/IST/IPLeiria Report on JPEG PCC Core Experiment 4.2: Sparse ConvolutionsPublication . Guarda, André; Rodrigues, Nuno; Pereira, FernandoContext and Objective: The current VM uses a dense 3D representation with dense convolutions Problem: Heavy in computation complexity, cannot encode a full point cloud at once; Underperforms for sparse point cloud Objective: Implement the DL models in the VM with a sparse tensor representation, and verify its performance
- IT/IST/IPLeiria Report on JPEG PCC Core Experiment 4.1: Attention ModelsPublication . Ghafari, Mohammadreza; Guarda, André; Rodrigues, Nuno; Pereira, FernandoContext and Objective: In the JPEG Pleno PC dataset, there are some PCs (e.g., sparse PCs) which are more ‘difficult’ to code and may benefit from improvements in the JPEG PCC VM DL coding model.
- IT/IST/IPLeiria Response to the Call for Evidence on JPEG Pleno Point Cloud CodingPublication . Rodrigues, Nuno M. M.; Pereira, Fernando; Guarda, AndréThis document proposes two scalable point cloud (PC) geometry codecs, submitted to the JPEG Call for Evidence on Point Cloud Coding (PCC).
- Deep Learning-based Compressed Domain Point Cloud ClassificationPublication . Seleem, Abdelrahman; Guarda, André; Rodrigues, Nuno; Pereira, FernandoThe JPEG Pleno PCC scope is a learning-based PC coding standard offering a singlestream, compact, compressed domain representation, targeting both human visualization, with significant compression efficiency improvement over PC coding standards in common use at equivalent subjective quality, as well as effective performance for PC processing and computer vision tasks.
- Constant Size Point Cloud Clustering: a Compact, Non-Overlapping SolutionPublication . Guarda, André F. R.; Rodrigues, Nuno M. M.; Pereira, FernandoPoint clouds have recently become a popular 3D representation model for many application domains, notably virtual and augmented reality. Since point cloud data is often very large, processing a point cloud may require that it be segmented into smaller clusters. For example, the input to deep learning-based methods like auto-encoders should be constant size point cloud clusters, which are ideally compact and non-overlapping. However, given the unorganized nature of point clouds, defining the specific data segments to code is not always trivial. This paper proposes a point cloud clustering algorithm which targets five main goals: i) clusters with a constant number of points; ii) compact clusters, i.e. with low dispersion; iii) non-overlapping clusters, i.e. not intersecting each other; iv) ability to scale with the number of points; and v) low complexity. After appropriate initialization, the proposed algorithm transfers points between neighboring clusters as a propagation wave, filling or emptying clusters until they achieve the same size. The proposed algorithm is unique since there is no other point cloud clustering method available in the literature offering the same clustering features for large point clouds at such low complexity