Search

Speed and consistency of target-shifting play a crucial role in human ability to perform complex tasks. Shifting our gaze between objects of interest quickly and consistently requires changes both in depth and direction. Gaze changes in depth are driven by slow, inconsistent vergence movements which rotate the eyes in opposite directions, while changes in direction are driven by ballistic, consistent movements called saccades, which rotate the eyes in the same direction. In the natural world, most of our eye movements are a combination of both types. While scientific consensus on the nature of saccades exists, vergence and combined movements remain less understood and agreed upon. We eschew the lack of scientific consensus in favor of proposing an operationalized computational model which predicts the speed of any type of gaze movement during target-shifting in 3D. To this end, we conduct a psychophysical study in a stereo VR environment to collect more than 12,000 gaze movement trials, analyze the temporal distribution of the observed gaze movements, and fit a probabilistic model to the data. We perform a series of objective measurements and user studies to validate the model. The results demonstrate its predictive accuracy, generalization, as well as applications for optimizing visual performance by altering content placement. Lastly, we leverage the model to measure differences in human target-changing time relative to the natural world, as well as suggest scene-aware projection depth. By incorporating the complexities and randomness of human oculomotor control, we hope this research will support new behavior-aware metrics for VR/AR display design, interface layout, and gaze-contingent rendering.

Low dynamic range (LDR) cameras cannot deal with wide dynamic range inputs, frequently leading to local overexposure issues. We present a learning-based system to reduce these artifacts without resorting to complex acquisition mechanisms like alternating exposures or costly processing that are typical of high dynamic range (HDR) imaging. We propose a transformer-based deep neural network (DNN) to infer the missing HDR details. In an ablation study, we show the importance of using a multiscale DNN and train it with the proper cost function to achieve state-of-the-art quality. To aid the reconstruction of the overexposed areas, our DNN takes a reference frame from the past as an additional input. This leverages the commonly occurring temporal instabilities of autoexposure to our advantage: since well-exposed details in the current frame may be overexposed in the future, we use reinforcement learning to train a reference frame selection DNN that decides whether to adopt the current frame as a future reference. Without resorting to alternating exposures, we obtain therefore a causal, HDR hallucination algorithm with potential application in common video acquisition settings. Our demo video can be found at https://drive.google.com/file/d/1-r12BKImLOYCLUoPzdebnMyNjJ4Rk360/view, Comment: The demo video can be found at https://drive.google.com/file/d/1-r12BKImLOYCLUoPzdebnMyNjJ4Rk360/view

We aim to ask and answer an essential question "how quickly do we react after observing a displayed visual target?" To this end, we present psychophysical studies that characterize the remarkable disconnect between human saccadic behaviors and spatial visual acuity. Building on the results of our studies, we develop a perceptual model to predict temporal gaze behavior, particularly saccadic latency, as a function of the statistics of a displayed image. Specifically, we implement a neurologically-inspired probabilistic model that mimics the accumulation of confidence that leads to a perceptual decision. We validate our model with a series of objective measurements and user studies using an eye-tracked VR display. The results demonstrate that our model prediction is in statistical alignment with real-world human behavior. Further, we establish that many sub-threshold image modifications commonly introduced in graphics pipelines may significantly alter human reaction timing, even if the differences are visually undetectable. Finally, we show that our model can serve as a metric to predict and alter reaction latency of users in interactive computer graphics applications, thus may improve gaze-contingent rendering, design of virtual experiences, and player performance in e-sports. We illustrate this with two examples: estimating competition fairness in a video game with two different team colors, and tuning display viewing distance to minimize player reaction time.

Previous blind or No Reference (NR) video quality assessment (VQA) models largely rely on features drawn from natural scene statistics (NSS), but under the assumption that the image statistics are stationary in the spatial domain. Several of these models are quite successful on standard pictures. However, in Virtual Reality (VR) applications, foveated video compression is regaining attention, and the concept of space-variant quality assessment is of interest, given the availability of increasingly high spatial and temporal resolution contents and practical ways of measuring gaze direction. Distortions from foveated video compression increase with increased eccentricity, implying that the natural scene statistics are space-variant. Towards advancing the development of foveated compression / streaming algorithms, we have devised a no-reference (NR) foveated video quality assessment model, called FOVQA, which is based on new models of space-variant natural scene statistics (NSS) and natural video statistics (NVS). Specifically, we deploy a space-variant generalized Gaussian distribution (SV-GGD) model and a space-variant asynchronous generalized Gaussian distribution (SV-AGGD) model of mean subtracted contrast normalized (MSCN) coefficients and products of neighboring MSCN coefficients, respectively. We devise a foveated video quality predictor that extracts radial basis features, and other features that capture perceptually annoying rapid quality fall-offs. We find that FOVQA achieves state-of-the-art (SOTA) performance on the new 2D LIVE-FBT-FCVR database, as compared with other leading FIQA / VQA models. we have made our implementation of FOVQA available at: http://live.ece.utexas.edu/research/Quality/FOVQA.zip.

Virtual Reality is regaining attention due to recent advancements in hardware technology. Immersive images / videos are becoming widely adopted to carry omnidirectional visual information. However, due to the requirements for higher spatial and temporal resolution of real video data, immersive videos require significantly larger bandwidth consumption. To reduce stresses on bandwidth, foveated video compression is regaining popularity, whereby the space-variant spatial resolution of the retina is exploited. Towards advancing the progress of foveated video compression, we propose a full reference (FR) foveated image quality assessment algorithm, which we call foveated entropic differencing (FED), which employs the natural scene statistics of bandpass responses by applying differences of local entropies weighted by a foveation-based error sensitivity function. We evaluate the proposed algorithm by measuring the correlations of the predictions that FED makes against human judgements on the newly created 2D and 3D LIVE-FBT-FCVR databases for Virtual Reality (VR). The performance of the proposed algorithm yields state-of-the-art as compared with other existing full reference algorithms. Software for FED has been made available at: http://live.ece.utexas.edu/research/Quality/FED.zip

With the development of higher resolution contents and displays, its significant volume poses significant challenges to the goals of acquiring, transmitting, compressing, and displaying high-quality video content. In this paper, we propose a new deep learning video compression architecture that does not require motion estimation, which is the most expensive element of modern hybrid video compression codecs like H.264 and HEVC. Our framework exploits the regularities inherent to video motion, which we capture by using displaced frame differences as video representations to train the neural network. In addition, we propose a new space-time reconstruction network based on both an LSTM model and a UNet model, which we call LSTM-UNet. The new video compression framework has three components: a Displacement Calculation Unit (DCU), a Displacement Compression Network (DCN), and a Frame Reconstruction Network (FRN). The DCU removes the need for motion estimation found in hybrid codecs and is less expensive. In the DCN, an RNN-based network is utilized to compress displaced frame differences as well as retain temporal information between frames. The LSTM-UNet is used in the FRN to learn space-time differential representations of videos. Our experimental results show that our compression model, which we call the MOtionless VIdeo Codec (MOVI-Codec), learns how to efficiently compress videos without computing motion. Our experiments show that MOVI-Codec outperforms the Low-Delay P veryfast setting of the video coding standard H.264 and exceeds the performance of the modern global standard HEVC codec, using the same setting, as measured by MS-SSIM, especially on higher resolution videos. In addition, our network outperforms the latest H.266 (VVC) codec at higher bitrates, when assessed using MS-SSIM, on high-resolution videos.

Disentanglement learning is crucial for obtaining disentangled representations and controllable generation. Current disentanglement methods face several inherent limitations: difficulty with high-resolution images, primarily focusing on learning disentangled representations, and non-identifiability due to the unsupervised setting. To alleviate these limitations, we design new architectures and loss functions based on StyleGAN (Karras et al., 2019), for semi-supervised high-resolution disentanglement learning. We create two complex high-resolution synthetic datasets for systematic testing. We investigate the impact of limited supervision and find that using only 0.25%~2.5% of labeled data is sufficient for good disentanglement on both synthetic and real datasets. We propose new metrics to quantify generator controllability, and observe there may exist a crucial trade-off between disentangled representation learning and controllable generation. We also consider semantic fine-grained image editing to achieve better generalization to unseen images., Comment: ICML 2020, 21 pages. Project page: https://sites.google.com/nvidia.com/semi-stylegan

In this paper, we propose algorithms which preserve energy in empirical mode decomposition (EMD), generating finite $n$ number of band limited Intrinsic Mode Functions (IMFs). In the first energy preserving EMD (EPEMD) algorithm, a signal is decomposed into linearly independent (LI), non orthogonal yet energy preserving (LINOEP) IMFs and residue (EPIMFs). It is shown that a vector in an inner product space can be represented as a sum of LI and non orthogonal vectors in such a way that Parseval's type property is satisfied. From the set of $n$ IMFs, through Gram-Schmidt orthogonalization method (GSOM), $n!$ set of orthogonal functions can be obtained. In the second algorithm, we show that if the orthogonalization process proceeds from lowest frequency IMF to highest frequency IMF, then the GSOM yields functions which preserve the properties of IMFs and the energy of a signal. With the Hilbert transform, these IMFs yield instantaneous frequencies and amplitudes as functions of time that reveal the imbedded structures of a signal. The instantaneous frequencies and square of amplitudes as functions of time produce a time-frequency-energy distribution, referred as the Hilbert spectrum, of a signal. Simulations have been carried out for the analysis of various time series and real life signals to show comparison among IMFs produced by EMD, EPEMD, ensemble EMD and multivariate EMD algorithms. Simulation results demonstrate the power of this proposed method., Comment: 23 pages, 25 figures

Since many decades, there is a general perception in literature that the Fourier methods are not suitable for the analysis of nonlinear and nonstationary data. In this paper, we propose a Fourier Decomposition Method (FDM) and demonstrate its efficacy for the analysis of nonlinear (i.e. data generated by nonlinear systems) and nonstationary time series. The proposed FDM decomposes any data into a small number of `Fourier intrinsic band functions' (FIBFs). The FDM presents a generalized Fourier expansion with variable amplitudes and frequencies of a time series by the Fourier method itself. We propose an idea of zero-phase filter bank based multivariate FDM (MFDM) algorithm, for the analysis of multivariate nonlinear and nonstationary time series, from the FDM. We also present an algorithm to obtain cutoff frequencies for MFDM. The MFDM algorithm is generating finite number of band limited multivariate FIBFs (MFIBFs). The MFDM preserves some intrinsic physical properties of the multivariate data, such as scale alignment, trend and instantaneous frequency. The proposed methods produce the results in a time-frequency-energy distribution that reveal the intrinsic structures of a data. Simulations have been carried out and comparison is made with the Empirical Mode Decomposition (EMD) methods in the analysis of various simulated as well as real life time series, and results show that the proposed methods are powerful tools for analyzing and obtaining the time-frequency-energy representation of any data., Comment: 14 Pages, 18 Figures

It is well known that, in any inner product space, a set of linearly independent (LI) vectors can be transformed to a set of orthogonal vectors, spanning the same space, by the Gram-Schmidt Orthogonalization Method (GSOM). In this paper, we propose a transformation from a set of LI vectors to a set of LI non orthogonal yet energy (square of the norm) preserving (LINOEP) vectors in an inner product space and we refer it as LINOEP method. We also show that there are various solutions to preserve the square of the norm., Comment: 6 pages, 2 figures

Blue noise sampling has proved useful for many graphics applications, but remains underexplored in high-dimensional spaces due to the difficulty of generating distributions and proving properties about them. We present a blue noise sampling method with good quality and performance across different dimensions. The method, spoke-dart sampling, shoots rays from prior samples and selects samples from these rays. It combines the advantages of two major high-dimensional sampling methods: the locality of advancing front with the dimensionality-reduction of hyperplanes, specifically line sampling. We prove that the output sampling is saturated with high probability, with bounds on distances between pairs of samples and between any domain point and its nearest sample. We demonstrate spoke-dart applications for approximate Delaunay graph construction, global optimization, and robotic motion planning. Both the blue-noise quality of the output distribution and the adaptability of the intermediate processes of our method are useful in these applications., Comment: 19 pages, 22 figures

We present Piko, a framework for designing, optimizing, and retargeting implementations of graphics pipelines on multiple architectures. Piko programmers express a graphics pipeline by organizing the computation within each stage into spatial bins and specifying a scheduling preference for these bins. Our compiler, Pikoc, compiles this input into an optimized implementation targeted to a massively-parallel GPU or a multicore CPU. Piko manages work granularity in a programmable and flexible manner, allowing programmers to build load-balanced parallel pipeline implementations, to exploit spatial and producer-consumer locality in a pipeline implementation, and to explore tradeoffs between these considerations. We demonstrate that Piko can implement a wide range of pipelines, including rasterization, Reyes, ray tracing, rasterization/ray tracing hybrid, and deferred rendering. Piko allows us to implement efficient graphics pipelines with relative ease and to quickly explore design alternatives by modifying the spatial binning configurations and scheduling preferences for individual stages, all while delivering real-time performance that is within a factor six of state-of-the-art rendering systems., Comment: 13 pages, updated for 2015

We formalize the notion of sampling a function using k-d darts. A k-d dart is a set of independent, mutually orthogonal, k-dimensional subspaces called k-d flats. Each dart has d choose k flats, aligned with the coordinate axes for efficiency. We show that k-d darts are useful for exploring a function's properties, such as estimating its integral, or finding an exemplar above a threshold. We describe a recipe for converting an algorithm from point sampling to k-d dart sampling, assuming the function can be evaluated along a k-d flat. We demonstrate that k-d darts are more efficient than point-wise samples in high dimensions, depending on the characteristics of the sampling domain: e.g. the subregion of interest has small volume and evaluating the function along a flat is not too expensive. We present three concrete applications using line darts (1-d darts): relaxed maximal Poisson-disk sampling, high-quality rasterization of depth-of-field blur, and estimation of the probability of failure from a response surface for uncertainty quantification. In these applications, line darts achieve the same fidelity output as point darts in less time. We also demonstrate the accuracy of higher dimensional darts for a volume estimation problem. For Poisson-disk sampling, we use significantly less memory, enabling the generation of larger point clouds in higher dimensions., Comment: 19 pages 16 figures