Image Coding for Machines via Feature-Preserving Rate-Distortion Optimization

Many images and videos are primarily processed by computer vision algorithms, involving only occasional human inspection. When this content requires compression before processing, e.g., in distributed applications, coding methods must optimize for both visual quality and downstream task performance. We first show that, given the features obtained from the original and the decoded images, an approach to reduce the effect of compression on a task loss is to perform rate-distortion optimization (RDO) using the distance between features as a distortion metric. However, optimizing directly such a rate-distortion trade-off requires an iterative workflow of encoding, decoding, and feature evaluation for each coding parameter, which is computationally impractical. We address this problem by simplifying the RDO formulation to make the distortion term computable using block-based encoders. We first apply Taylor's expansion to the feature extractor, recasting the feature distance as a quadratic metric with the Jacobian matrix of the neural network. Then, we replace the linearized metric with a block-wise approximation, which we call input-dependent squared error (IDSE). To reduce computational complexity, we approximate IDSE using Jacobian sketches. The resulting loss can be evaluated block-wise in the transform domain and combined with the sum of squared errors (SSE) to address both visual quality and computer vision performance. Simulations with AVC across multiple feature extractors and downstream neural networks show up to 10% bit-rate savings for the same computer vision accuracy compared to RDO based on SSE, with no decoder complexity overhead and just a 7% encoder complexity increase.

Towards joint graph learning and sampling set selection from data

We explore the problem of sampling graph signals in scenarios where the graph structure is not predefined and must be inferred from data. In this scenario, existing approaches rely on a two-step process, where a graph is learned first, followed by sampling. More generally, graph learning and graph signal sampling have been studied as two independent problems in the literature. This work provides a foundational step towards jointly optimizing the graph structure and sampling set. Our main contribution, Vertex Importance Sampling (VIS), is to show that the sampling set can be effectively determined from the vertex importance (node weights) obtained from graph learning. We further propose Vertex Importance Sampling with Repulsion (VISR), a greedy algorithm where spatially -separated "important" nodes are selected to ensure better reconstruction. Empirical results on simulated data show that sampling using VIS and VISR leads to competitive reconstruction performance and lower complexity than the conventional two-step approach of graph learning followed by graph sampling.

Towards joint graph and sampling set selection from data

We explore the problem of sampling graph signals in scenarios where the graph structure is not predefined and must be inferred from data. In this scenario, existing approaches rely on a two-step process, where a graph is learned first, followed by sampling. More generally, graph learning and graph signal sampling have been studied as two independent problems in the literature. This work provides a foundational step towards jointly optimizing the graph structure and sampling set. Our main contribution, Vertex Importance Sampling (VIS), is to show that the sampling set can be effectively determined from the vertex importance (node weights) obtained from graph learning. We further propose Vertex Importance Sampling with Repulsion (VISR), a greedy algorithm where spatially -separated "important" nodes are selected to ensure better reconstruction. Empirical results on simulated data show that sampling using VIS and VISR leads to competitive reconstruction performance and lower complexity than the conventional two-step approach of graph learning followed by graph sampling.

Color-Guided Flying Pixel Correction in Depth Images

We present a novel method to correct flying pixels within data captured by Time-of-flight (ToF) sensors. Flying pixel (FP) artifacts occur when signals from foreground and background objects reach the same sensor pixel, leading to a confident yet incorrect depth estimation in space - floating between two objects. Commercial RGB-D cameras have a complementary setup consisting of ToF sensors to capture depth in addition to RGB cameras. We propose a novel method to correct FPs by leveraging the aligned RGB and depth image in such RGB-D cameras to estimate the true depth values of FPs. Our method defines a 3D neighborhood around each point, representing a "field of view" that mirrors the acquisition process of ToF cameras. We propose a two-step iterative correction algorithm in which the FPs are first identified. Then, we estimate the true depth value of FPs by solving a least-squares optimization problem. Experimental results show that our proposed algorithm estimates the depth value of FPs as accurately as other algorithms in the literature.

Graph-based Scalable Sampling of 3D Point Cloud Attributes

3D Point clouds (PCs) are commonly used to represent 3D scenes. They can have millions of points, making subsequent downstream tasks such as compression and streaming computationally expensive. PC sampling (selecting a subset of points) can be used to reduce complexity. Existing PC sampling algorithms focus on preserving geometry features and often do not scale to handle large PCs. In this work, we develop scalable graph-based sampling algorithms for PC color attributes, assuming the full geometry is available. Our sampling algorithms are optimized for a signal reconstruction method that minimizes the graph Laplacian quadratic form. We first develop a global sampling algorithm that can be applied to PCs with millions of points by exploiting sparsity and sampling rate adaptive parameter selection. Further, we propose a block-based sampling strategy where each block is sampled independently. We show that sampling the corresponding sub-graphs with optimally chosen self-loop weights (node weights) will produce a sampling set that approximates the results of global sampling while reducing complexity by an order of magnitude. Our empirical results on two large PC datasets show that our algorithms outperform the existing fast PC subsampling techniques (uniform and geometry feature preserving random sampling) by 2dB. Our algorithm is up to 50 times faster than existing graph signal sampling algorithms while providing better reconstruction accuracy. Finally, we illustrate the efficacy of PC attribute sampling within a compression scenario, showing that pre-compression sampling of PC attributes can lower the bitrate by 11% while having minimal effect on reconstruction.

Graph-Based Signal Sampling with Adaptive Subspace Reconstruction for Spatially-Irregular Sensor Data

Choosing an appropriate frequency definition and norm is critical in graph signal sampling and reconstruction. Most previous works define frequencies based on the spectral properties of the graph and use the same frequency definition and $\ell_2$-norm for optimization for all sampling sets. Our previous work demonstrated that using a sampling set-adaptive norm and frequency definition can address challenges in classical bandlimited approximation, particularly with model mismatches and irregularly distributed data. In this work, we propose a method for selecting sampling sets tailored to the sampling set adaptive GFT-based interpolation. When the graph models the inverse covariance of the data, we show that this adaptive GFT enables localizing the bandlimited model mismatch error to high frequencies, and the spectral folding property allows us to track this error in reconstruction. Based on this, we propose a sampling set selection algorithm to minimize the worst-case bandlimited model mismatch error. We consider partitioning the sensors in a sensor network sampling a continuous spatial process as an application. Our experiments show that sampling and reconstruction using sampling set adaptive GFT significantly outperform methods that used fixed GFTs and bandwidth-based criterion.

Fast DCT+: A Family of Fast Transforms Based on Rank-One Updates of the Path Graph

This paper develops fast graph Fourier transform (GFT) algorithms with O(n log n) runtime complexity for rank-one updates of the path graph. We first show that several commonly-used audio and video coding transforms belong to this class of GFTs, which we denote by DCT+. Next, starting from an arbitrary generalized graph Laplacian and using rank-one perturbation theory, we provide a factorization for the GFT after perturbation. This factorization is our central result and reveals a progressive structure: we first apply the unperturbed Laplacian's GFT and then multiply the result by a Cauchy matrix. By specializing this decomposition to path graphs and exploiting the properties of Cauchy matrices, we show that Fast DCT+ algorithms exist. We also demonstrate that progressivity can speed up computations in applications involving multiple transforms related by rank-one perturbations (e.g., video coding) when combined with pruning strategies. Our results can be extended to other graphs and rank-k perturbations. Runtime analyses show that Fast DCT+ provides computational gains over the naive method for graph sizes larger than 64, with runtime approximately equal to that of 8 DCTs.

Feature-Preserving Rate-Distortion Optimization in Image Coding for Machines

With the increasing number of images and videos consumed by computer vision algorithms, compression methods are evolving to consider both perceptual quality and performance in downstream tasks. Traditional codecs can tackle this problem by performing rate-distortion optimization (RDO) to minimize the distance at the output of a feature extractor. However, neural network non-linearities can make the rate-distortion landscape irregular, leading to reconstructions with poor visual quality even for high bit rates. Moreover, RDO decisions are made block-wise, while the feature extractor requires the whole image to exploit global information. In this paper, we address these limitations in three steps. First, we apply Taylor's expansion to the feature extractor, recasting the metric as an input-dependent squared error involving the Jacobian matrix of the neural network. Second, we make a localization assumption to compute the metric block-wise. Finally, we use randomized dimensionality reduction techniques to approximate the Jacobian. The resulting expression is monotonic with the rate and can be evaluated in the transform domain. Simulations with AVC show that our approach provides bit-rate savings while preserving accuracy in downstream tasks with less complexity than using the feature distance directly.

Full reference point cloud quality assessment using support vector regression

Point clouds are a general format for representing realistic 3D objects in diverse 3D applications. Since point clouds have large data sizes, developing efficient point cloud compression methods is crucial. However, excessive compression leads to various distortions, which deteriorates the point cloud quality perceived by end users. Thus, establishing reliable point cloud quality assessment (PCQA) methods is essential as a benchmark to develop efficient compression methods. This paper presents an accurate full-reference point cloud quality assessment (FR-PCQA) method called full-reference quality assessment using support vector regression (FRSVR) for various types of degradations such as compression distortion, Gaussian noise, and down-sampling. The proposed method demonstrates accurate PCQA by integrating five FR-based metrics covering various types of errors (e.g., considering geometric distortion, color distortion, and point count) using support vector regression (SVR). Moreover, the proposed method achieves a superior trade-off between accuracy and calculation speed because it includes only the calculation of these five simple metrics and SVR, which can perform fast prediction. Experimental results with three types of open datasets show that the proposed method is more accurate than conventional FR-PCQA methods. In addition, the proposed method is faster than state-of-the-art methods that utilize complicated features such as curvature and multi-scale features. Thus, the proposed method provides excellent performance in terms of the accuracy of PCQA and processing speed. Our method is available from https://github.com/STAC-USC/FRSVR-PCQA.

Understanding Encoder-Decoder Structures in Machine Learning Using Information Measures

We present new results to model and understand the role of encoder-decoder design in machine learning (ML) from an information-theoretic angle. We use two main information concepts, information sufficiency (IS) and mutual information loss (MIL), to represent predictive structures in machine learning. Our first main result provides a functional expression that characterizes the class of probabilistic models consistent with an IS encoder-decoder latent predictive structure. This result formally justifies the encoder-decoder forward stages many modern ML architectures adopt to learn latent (compressed) representations for classification. To illustrate IS as a realistic and relevant model assumption, we revisit some known ML concepts and present some interesting new examples: invariant, robust, sparse, and digital models. Furthermore, our IS characterization allows us to tackle the fundamental question of how much performance (predictive expressiveness) could be lost, using the cross entropy risk, when a given encoder-decoder architecture is adopted in a learning setting. Here, our second main result shows that a mutual information loss quantifies the lack of expressiveness attributed to the choice of a (biased) encoder-decoder ML design. Finally, we address the problem of universal cross-entropy learning with an encoder-decoder design where necessary and sufficiency conditions are established to meet this requirement. In all these results, Shannon's information measures offer new interpretations and explanations for representation learning.