arXiv subjects

Antonio Ortega

Publications and source records attributed to Antonio Ortega.

At least 73 records · Page 4Linked to original sources

arXiv2023

Rate-Distortion Optimization With Alternative References For UGC Video Compression

User generated content (UGC) refers to videos that are uploaded by users and shared over the Internet. UGC may have low quality due to noise and previous compression. When re-encoding UGC for streaming or downloading, a traditional video coding pipeline will perform rate-distortion (RD) optimization to choose coding parameters. However, in the UGC video coding case, since the input is not pristine, quality ``saturation'' (or even degradation) can be observed, i.e., increased bitrate only leads to improved representation of coding artifacts and noise present in the UGC input. In this paper, we study the saturation problem in UGC compression, where the goal is to identify and avoid during encoding, the coding parameters and rates that lead to quality saturation. We proposed a geometric criterion for saturation detection that works with rate-distortion optimization, and only requires a few frames from the UGC video. In addition, we show how to combine the proposed saturation detection method with existing video coding systems that implement rate-distortion optimization for efficient compression of UGC videos.

eess.IV↗

arXiv2023

Image Coding via Perceptually Inspired Graph Learning

Most codec designs rely on the mean squared error (MSE) as a fidelity metric in rate-distortion optimization, which allows to choose the optimal parameters in the transform domain but may fail to reflect perceptual quality. Alternative distortion metrics, such as the structural similarity index (SSIM), can be computed only pixel-wise, so they cannot be used directly for transform-domain bit allocation. Recently, the irregularity-aware graph Fourier transform (IAGFT) emerged as a means to include pixel-wise perceptual information in the transform design. This paper extends this idea by also learning a graph (and corresponding transform) for sets of blocks that share similar perceptual characteristics and are observed to differ statistically, leading to different learned graphs. We demonstrate the effectiveness of our method with both SSIM- and saliency-based criteria. We also propose a framework to derive separable transforms, including separable IAGFTs. An empirical evaluation based on the 5th CLIC dataset shows that our approach achieves improvements in terms of MS-SSIM with respect to existing methods.

eess.IV↗

arXiv2023

Two Channel Filter Banks on Arbitrary Graphs with Positive Semi Definite Variation Operators

We propose novel two-channel filter banks for signals on graphs. Our designs can be applied to arbitrary graphs, given a positive semi definite variation operator, while using arbitrary vertex partitions for downsampling. The proposed generalized filter banks (GFBs) also satisfy several desirable properties including perfect reconstruction and critical sampling, while having efficient implementations. Our results generalize previous approaches that were only valid for the normalized Laplacian of bipartite graphs. Our approach is based on novel graph Fourier transforms (GFTs) given by the generalized eigenvectors of the variation operator. These GFTs are orthogonal in an alternative inner product space which depends on the downsampling and variation operators. Our key theoretical contribution is showing that the spectral folding property of the normalized Laplacian of bipartite graphs, at the core of bipartite filter bank theory, can be generalized for the proposed GFT if the inner product matrix is chosen properly. In addition, we study vertex domain and spectral domain properties of GFBs and illustrate their probabilistic interpretation using Gaussian graphical models. While GFBs can be defined given any choice of a vertex partition for downsampling, we propose an algorithm to optimize these partitions with a criterion that favors balanced partitions with large graph cuts, which are shown to lead to efficient and stable GFB implementations. Our numerical experiments show that partition-optimized GFBs can be implemented efficiently on 3D point clouds with hundreds of thousands of points (nodes), while also improving the color signal representation quality over competing state-of-the-art approaches.

eess.SP↗

arXiv2022

Motion estimation and filtered prediction for dynamic point cloud attribute compression

In point cloud compression, exploiting temporal redundancy for inter predictive coding is challenging because of the irregular geometry. This paper proposes an efficient block-based inter-coding scheme for color attribute compression. The scheme includes integer-precision motion estimation and an adaptive graph based in-loop filtering scheme for improved attribute prediction. The proposed block-based motion estimation scheme consists of an initial motion search that exploits geometric and color attributes, followed by a motion refinement that only minimizes color prediction error. To further improve color prediction, we propose a vertex-domain low-pass graph filtering scheme that can adaptively remove noise from predictors computed from motion estimation with different accuracy. Our experiments demonstrate significant coding gain over state-of-the-art coding methods.

cs.CV↗

arXiv2022

Toward a Geometrical Understanding of Self-supervised Contrastive Learning

Self-supervised learning (SSL) is currently one of the premier techniques to create data representations that are actionable for transfer learning in the absence of human annotations. Despite their success, the underlying geometry of these representations remains elusive, which obfuscates the quest for more robust, trustworthy, and interpretable models. In particular, mainstream SSL techniques rely on a specific deep neural network architecture with two cascaded neural networks: the encoder and the projector. When used for transfer learning, the projector is discarded since empirical results show that its representation generalizes more poorly than the encoder's. In this paper, we investigate this curious phenomenon and analyze how the strength of the data augmentation policies affects the data embedding. We discover a non-trivial relation between the encoder, the projector, and the data augmentation strength: with increasingly larger augmentation policies, the projector, rather than the encoder, is more strongly driven to become invariant to the augmentations. It does so by eliminating crucial information about the data by learning to project it into a low-dimensional space, a noisy estimate of the data manifold tangent plane in the encoder representation. This analysis is substantiated through a geometrical perspective with theoretical and empirical results.

cs.LG↗

arXiv2022

The Geometry of Self-supervised Learning Models and its Impact on Transfer Learning

Self-supervised learning (SSL) has emerged as a desirable paradigm in computer vision due to the inability of supervised models to learn representations that can generalize in domains with limited labels. The recent popularity of SSL has led to the development of several models that make use of diverse training strategies, architectures, and data augmentation policies with no existing unified framework to study or assess their effectiveness in transfer learning. We propose a data-driven geometric strategy to analyze different SSL models using local neighborhoods in the feature space induced by each. Unlike existing approaches that consider mathematical approximations of the parameters, individual components, or optimization landscape, our work aims to explore the geometric properties of the representation manifolds learned by SSL models. Our proposed manifold graph metrics (MGMs) provide insights into the geometric similarities and differences between available SSL models, their invariances with respect to specific augmentations, and their performances on transfer learning tasks. Our key findings are two fold: (i) contrary to popular belief, the geometry of SSL models is not tied to its training paradigm (contrastive, non-contrastive, and cluster-based); (ii) we can predict the transfer learning capability for a specific model based on the geometric properties of its semantic and augmentation manifolds.

cs.LG↗

arXiv2022

Compression of user generated content using denoised references

Video shared over the internet is commonly referred to as user generated content (UGC). UGC video may have low quality due to various factors including previous compression. UGC video is uploaded by users, and then it is re-encoded to be made available at various levels of quality. In a traditional video coding pipeline the encoder parameters are optimized to minimize a rate-distortion criterion, but when the input signal has low quality, this results in sub-optimal coding parameters optimized to preserve undesirable artifacts. In this paper we formulate the UGC compression problem as that of compression of a noisy/corrupted source. The noisy source coding theorem reveals that an optimal UGC compression system is comprised of optimal denoising of the UGC signal, followed by compression of the denoised signal. Since optimal denoising is unattainable and users may be against modification of their content, we propose encoding the UGC signal, and using denoised references only to compute distortion, so the encoding process can be guided towards perceptually better solutions. We demonstrate the effectiveness of the proposed strategy for JPEG compression of UGC images and videos.

eess.IV↗

arXiv2022

NNK-Means: Data summarization using dictionary learning with non-negative kernel regression

An increasing number of systems are being designed by gathering significant amounts of data and then optimizing the system parameters directly using the obtained data. Often this is done without analyzing the dataset structure. As task complexity, data size, and parameters all increase to millions or even billions, data summarization is becoming a major challenge. In this work, we investigate data summarization via dictionary learning~(DL), leveraging the properties of recently introduced non-negative kernel regression (NNK) graphs. Our proposed NNK-Means, unlike previous DL techniques, such as kSVD, learns geometric dictionaries with atoms that are representative of the input data space. Experiments show that summarization using NNK-Means can provide better class separation compared to linear and kernel versions of kMeans and kSVD. Moreover, NNK-Means is scalable, with runtime complexity similar to that of kMeans.

cs.LG↗

arXiv2022

Hybrid Model-based / Data-driven Graph Transform for Image Coding

Transform coding to sparsify signal representations remains crucial in an image compression pipeline. While the Karhunen-Loève transform (KLT) computed from an empirical covariance matrix $\bar{C}$ is theoretically optimal for a stationary process, in practice, collecting sufficient statistics from a non-stationary image to reliably estimate $\bar{C}$ can be difficult. In this paper, to encode an intra-prediction residual block, we pursue a hybrid model-based / data-driven approach: the first $K$ eigenvectors of a transform matrix are derived from a statistical model, e.g., the asymmetric discrete sine transform (ADST), for stability, while the remaining $N-K$ are computed from $\bar{C}$ for performance. The transform computation is posed as a graph learning problem, where we seek a graph Laplacian matrix minimizing a graphical lasso objective inside a convex cone sharing the first $K$ eigenvectors in a Hilbert space of real symmetric matrices. We efficiently solve the problem via augmented Lagrangian relaxation and proximal gradient (PG). Using WebP as a baseline image codec, experimental results show that our hybrid graph transform achieved better energy compaction than default discrete cosine transform (DCT) and better stability than KLT.

eess.IV↗

arXiv2022

Practical graph signal sampling with log-linear size scaling

Graph signal sampling is the problem of selecting a subset of representative graph vertices whose values can be used to interpolate missing values on the remaining graph vertices. Optimizing the choice of sampling set using concepts from experiment design can help minimize the effect of noise in the input signal. While many existing sampling set selection methods are computationally intensive because they require an eigendecomposition, existing eigendecompostion-free methods are still much slower than random sampling algorithms for large graphs. In this paper, through optimizing sampling sets towards the D-optimal objective from experiment design, we propose a sampling algorithm that has complexity comparable to random sampling algorithms, while reaching accuracy similar to existing eigendecomposition-free methods for a broad range of graph types.

eess.SP↗

arXiv2022

Fractional Motion Estimation for Point Cloud Compression

Motivated by the success of fractional pixel motion in video coding, we explore the design of motion estimation with fractional-voxel resolution for compression of color attributes of dynamic 3D point clouds. Our proposed block-based fractional-voxel motion estimation scheme takes into account the fundamental differences between point clouds and videos, i.e., the irregularity of the distribution of voxels within a frame and across frames. We show that motion compensation can benefit from the higher resolution reference and more accurate displacements provided by fractional precision. Our proposed scheme significantly outperforms comparable methods that only use integer motion. The proposed scheme can be combined with and add sizeable gains to state-of-the-art systems that use transforms such as Region Adaptive Graph Fourier Transform and Region Adaptive Haar Transform.

eess.IV↗

arXiv2022

Pre-demosaic Graph-based Light Field Image Compression

An unfocused plenoptic light field (LF) camera places an array of microlenses in front of an image sensor in order to separately capture different directional rays arriving at an image pixel. Using a conventional Bayer pattern, data captured at each pixel is a single color component (R, G or B).The sensed data then undergoes demosaicking (interpolation of RGB components per pixel) and conversion to an array of sub-aperture images (SAIs). In this paper, we propose a new LF image coding scheme based on graph lifting transform (GLT), where the acquired sensor data are coded in the original captured form without pre-processing. Specifically, we directly map raw sensed color data to the SAIs, resulting in sparsely distributed color pixels on 2D grids, and perform demosaicking at the receiver after decoding. To exploit spatial correlation among the sparse pixels, we propose a novel intra-prediction scheme, where the prediction kernel is determined according to the local gradient estimated from already coded neighboring pixel blocks. We then connect the pixels by forming a graph, modeling the prediction residuals statistically as a Gaussian Markov Random Field (GMRF). The optimal edge weights are computed via a graph learning method using a set of training SAIs. The residual data is encoded via low-complexity GLT. Experiments show that at high PSNRs -- important for archiving and instant storage scenarios -- our method outperformed significantly a conventional light field image coding scheme with demosaicking followed by High Efficiency Video Coding (HEVC).

eess.IV↗

arXiv2021

Channel redundancy and overlap in convolutional neural networks with channel-wise NNK graphs

Feature spaces in the deep layers of convolutional neural networks (CNNs) are often very high-dimensional and difficult to interpret. However, convolutional layers consist of multiple channels that are activated by different types of inputs, which suggests that more insights may be gained by studying the channels and how they relate to each other. In this paper, we first analyze theoretically channel-wise non-negative kernel (CW-NNK) regression graphs, which allow us to quantify the overlap between channels and, indirectly, the intrinsic dimension of the data representation manifold. We find that redundancy between channels is significant and varies with the layer depth and the level of regularization during training. Additionally, we observe that there is a correlation between channel overlap in the last convolutional layer and generalization performance. Our experimental results demonstrate that these techniques can lead to a better understanding of deep representations.

cs.LG↗

arXiv2021

Orthogonal Transforms for Signals on Directed Graphs

In this paper we consider the problem of defining transforms for signals on directed graphs, with a specific focus on defective graphs where the corresponding graph operator cannot be diagonalized. Our proposed method is based on the Schur decomposition and leads to a series of embedded invariant subspaces for which orthogonal basis are available. As compared to diffusion wavelets, our method is more flexible in the generation of subspaces, but these subspaces can only be approximately orthogonal.

eess.SP↗

arXiv2021

Learning Sparse Graph with Minimax Concave Penalty under Gaussian Markov Random Fields

This paper presents a convex-analytic framework to learn sparse graphs from data. While our problem formulation is inspired by an extension of the graphical lasso using the so-called combinatorial graph Laplacian framework, a key difference is the use of a nonconvex alternative to the $\ell_1$ norm to attain graphs with better interpretability. Specifically, we use the weakly-convex minimax concave penalty (the difference between the $\ell_1$ norm and the Huber function) which is known to yield sparse solutions with lower estimation bias than $\ell_1$ for regression problems. In our framework, the graph Laplacian is replaced in the optimization by a linear transform of the vector corresponding to its upper triangular part. Via a reformulation relying on Moreau's decomposition, we show that overall convexity is guaranteed by introducing a quadratic function to our cost function. The problem can be solved efficiently by the primal-dual splitting method, of which the admissible conditions for provable convergence are presented. Numerical examples show that the proposed method significantly outperforms the existing graph learning methods with reasonable CPU time.

eess.SP↗

arXiv2021

Estimating Active Cases of COVID-19

Having accurate and timely data on confirmed active COVID-19 cases is challenging, since it depends on testing capacity and the availability of an appropriate infrastructure to perform tests and aggregate their results. In this paper, we propose methods to estimate the number of active cases of COVID-19 from the official data (of confirmed cases and fatalities) and from survey data. We show that the latter is a viable option in countries with reduced testing capacity or suboptimal infrastructures.

physics.soc-ph↗

arXiv2021

Channel-Wise Early Stopping without a Validation Set via NNK Polytope Interpolation

State-of-the-art neural network architectures continue to scale in size and deliver impressive generalization results, although this comes at the expense of limited interpretability. In particular, a key challenge is to determine when to stop training the model, as this has a significant impact on generalization. Convolutional neural networks (ConvNets) comprise high-dimensional feature spaces formed by the aggregation of multiple channels, where analyzing intermediate data representations and the model's evolution can be challenging owing to the curse of dimensionality. We present channel-wise DeepNNK (CW-DeepNNK), a novel channel-wise generalization estimate based on non-negative kernel regression (NNK) graphs with which we perform local polytope interpolation on low-dimensional channels. This method leads to instance-based interpretability of both the learned data representations and the relationship between channels. Motivated by our observations, we use CW-DeepNNK to propose a novel early stopping criterion that (i) does not require a validation set, (ii) is based on a task performance metric, and (iii) allows stopping to be reached at different points for each channel. Our experiments demonstrate that our proposed method has advantages as compared to the standard criterion based on validation set performance.

cs.LG↗

arXiv2021

Cylindrical coordinates for LiDAR point cloud compression

We present an efficient voxelization method to encode the geometry and attributes of 3D point clouds obtained from autonomous vehicles. Due to the circular scanning trajectory of sensors, the geometry of LiDAR point clouds is inherently different from that of point clouds captured from RGBD cameras. Our method exploits these specific properties to representing points in cylindrical coordinates instead of conventional Cartesian coordinates. We demonstrate thatRegion Adaptive Hierarchical Transform (RAHT) can be extended to this setting, leading to attribute encoding based on a volumetric partition in cylindrical coordinates. Experimental results show that our proposed voxelization outperforms conventional approaches based on Cartesian coordinates for this type of data. We observe a significant improvement in attribute coding performance with 5-10%reduction in bitrate and octree representation with 35-45% reduction in bits.

eess.IV↗