arXiv subjects

Zhan Ma

Publications and source records attributed to Zhan Ma.

At least 91 records · Page 5Linked to original sources

arXiv2021

Dynamical Clustering Interrupts Motility Induced Phase Separation in Chiral Active Brownian Particles

One of the most intriguing phenomena in active matter has been the gas-liquid like motility induced phase separation (MIPS) observed in repulsive active particles. However, experimentally no particle can be a perfect sphere, and the asymmetric shape, mass distribution or catalysis coating can induce an active torque on the particle, which makes it a chiral active particle. Here using computer simulations and dynamic mean-field theory, we demonstrate that the large enough torque of circle active Brownian particles (cABPs) in two dimensions generates a dynamical clustering state interrupting the conventional MIPS. Multiple clusters arise from the combination of the conventional MIPS cohesion, and the circulating current caused disintegration. The non-vanishing current in non-equilibrium steady states microscopically originates from the motility ``relieved'' by automatic rotation, which breaks the detailed balance at the continuum level. This suggests that no equilibrium-like phase separation theory can be constructed for chiral active colloids even with tiny active torque, in which no visible collective motion exists. This mechanism also sheds light on the understanding of dynamic clusters observed in a variety of active matter systems.

cond-mat.soft↗

arXiv2021

Transformer-based Image Compression

A Transformer-based Image Compression (TIC) approach is developed which reuses the canonical variational autoencoder (VAE) architecture with paired main and hyper encoder-decoders. Both main and hyper encoders are comprised of a sequence of neural transformation units (NTUs) to analyse and aggregate important information for more compact representation of input image, while the decoders mirror the encoder-side operations to generate pixel-domain image reconstruction from the compressed bitstream. Each NTU is consist of a Swin Transformer Block (STB) and a convolutional layer (Conv) to best embed both long-range and short-range information; In the meantime, a casual attention module (CAM) is devised for adaptive context modeling of latent features to utilize both hyper and autoregressive priors. The TIC rivals with state-of-the-art approaches including deep convolutional neural networks (CNNs) based learnt image coding (LIC) methods and handcrafted rules-based intra profile of recently-approved Versatile Video Coding (VVC) standard, and requires much less model parameters, e.g., up to 45% reduction to leading-performance LIC.

eess.IV↗

arXiv2021

Which One is Better: Assessing Objective Metrics for Point Cloud Compression

Point cloud compression (PCC) has made remarkable achievement in recent years. In the mean time, point cloud quality assessment (PCQA) also realize gratifying development. Some recently emerged metrics present robust performance on public point cloud assessment databases. However, these metrics have not been evaluated specifically for PCC to verify whether they exhibit consistent performance with the subjective perception. In this paper, we establish a new dataset for compression evaluation first, which contains 175 compressed point clouds in total, deriving from 7 compression algorithms with 5 compression levels. Then leveraging the proposed dataset, we evaluate the performance of the existing PCQA metrics in terms of different compression types. The results demonstrate some deficiencies of existing metrics in compression evaluation.

eess.IV↗

arXiv2021

End-to-end Neural Video Coding Using a Compound Spatiotemporal Representation

Recent years have witnessed rapid advances in learnt video coding. Most algorithms have solely relied on the vector-based motion representation and resampling (e.g., optical flow based bilinear sampling) for exploiting the inter frame redundancy. In spite of the great success of adaptive kernel-based resampling (e.g., adaptive convolutions and deformable convolutions) in video prediction for uncompressed videos, integrating such approaches with rate-distortion optimization for inter frame coding has been less successful. Recognizing that each resampling solution offers unique advantages in regions with different motion and texture characteristics, we propose a hybrid motion compensation (HMC) method that adaptively combines the predictions generated by these two approaches. Specifically, we generate a compound spatiotemporal representation (CSTR) through a recurrent information aggregation (RIA) module using information from the current and multiple past frames. We further design a one-to-many decoder pipeline to generate multiple predictions from the CSTR, including vector-based resampling, adaptive kernel-based resampling, compensation mode selection maps and texture enhancements, and combines them adaptively to achieve more accurate inter prediction. Experiments show that our proposed inter coding system can provide better motion-compensated prediction and is more robust to occlusions and complex motions. Together with jointly trained intra coder and residual coder, the overall learnt hybrid coder yields the state-of-the-art coding efficiency in low-delay scenario, compared to the traditional H.264/AVC and H.265/HEVC, as well as recently published learning-based methods, in terms of both PSNR and MS-SSIM metrics.

eess.IV↗

arXiv2021

ANT: Learning Accurate Network Throughput for Better Adaptive Video Streaming

Adaptive Bit Rate (ABR) decision plays a crucial role for ensuring satisfactory Quality of Experience (QoE) in video streaming applications, in which past network statistics are mainly leveraged for future network bandwidth prediction. However, most algorithms, either rules-based or learning-driven approaches, feed throughput traces or classified traces based on traditional statistics (i.e., mean/standard deviation) to drive ABR decision, leading to compromised performances in specific scenarios. Given the diverse network connections (e.g., WiFi, cellular and wired link) from time to time, this paper thus proposes to learn the ANT (a.k.a., Accurate Network Throughput) model to characterize the full spectrum of network throughput dynamics in the past for deriving the proper network condition associated with a specific cluster of network throughput segments (NTS). Each cluster of NTS is then used to generate a dedicated ABR model, by which we wish to better capture the network dynamics for diverse connections. We have integrated the ANT model with existing reinforcement learning (RL)-based ABR decision engine, where different ABR models are applied to respond to the accurate network sensing for better rate decision. Extensive experiment results show that our approach can significantly improve the user QoE by 65.5% and 31.3% respectively, compared with the state-of-the-art Pensive and Oboe, across a wide range of network scenarios.

cs.MM↗

arXiv2021

Data-driven Coarse-grained Modeling of Non-equilibrium Systems

Modeling a high-dimensional Hamiltonian system in reduced dimensions with respect to coarse-grained (CG) variables can greatly reduce computational cost and enable efficient bottom-up prediction of main features of the system for many applications. However, it usually experiences significantly altered dynamics due to loss of degrees of freedom upon coarse-graining. To establish CG models that can faithfully preserve dynamics, previous efforts mainly focused on equilibrium systems. In contrast, various soft matter systems are known out of equilibrium. Therefore, the present work concerns non-equilibrium systems and enables accurate and efficient CG modeling that preserves non-equilibrium dynamics and is generally applicable to any non-equilibrium process and any observable of interest. To this end, the dynamic equation of a CG variable is built in the form of the non-stationary generalized Langevin equation (nsGLE) to account for the dependence of non-equilibrium processes on the initial conditions, where the two-time memory kernel is determined from the data of the two-time auto-correlation function of the non-equilibrium trajectory-averaged observable of interest. By embedding the non-stationary non-Markovian process in an extended stochastic framework, an explicit form of the non-stationary random noise in the nsGLE is introduced, and the cost is significantly reduced for solving the nsGLE to predict the non-equilibrium dynamics of the CG variable. To prove and exploit the equivalence of the nsGLE and extended dynamics, the memory kernel is parameterized in a two-time exponential expansion. A data-driven hybrid optimization process is proposed for the parameterization, a non-convex and high-dimensional optimization problem.

cs.CE↗

arXiv2021

Learning End-to-End Lossy Image Compression: A Benchmark

Image compression is one of the most fundamental techniques and commonly used applications in the image and video processing field. Earlier methods built a well-designed pipeline, and efforts were made to improve all modules of the pipeline by handcrafted tuning. Later, tremendous contributions were made, especially when data-driven methods revitalized the domain with their excellent modeling capacities and flexibility in incorporating newly designed modules and constraints. Despite great progress, a systematic benchmark and comprehensive analysis of end-to-end learned image compression methods are lacking. In this paper, we first conduct a comprehensive literature survey of learned image compression methods. The literature is organized based on several aspects to jointly optimize the rate-distortion performance with a neural network, i.e., network architecture, entropy model and rate control. We describe milestones in cutting-edge learned image-compression methods, review a broad range of existing works, and provide insights into their historical development routes. With this survey, the main challenges of image compression methods are revealed, along with opportunities to address the related issues with recent advanced learning methods. This analysis provides an opportunity to take a further step towards higher-efficiency image compression. By introducing a coarse-to-fine hyperprior model for entropy estimation and signal reconstruction, we achieve improved rate-distortion performance, especially on high-resolution images. Extensive benchmark experiments demonstrate the superiority of our model in rate-distortion performance and time complexity on multi-core CPUs and GPUs. Our project website is available at https://huzi96.github.io/compression-bench.html.

eess.IV↗

arXiv2021

Transfer Learning of Memory Kernels in Coarse-grained Modeling

The present work concerns the transferability of coarse-grained (CG) modeling in reproducing the dynamic properties of the reference atomistic systems across a range of parameters. In particular, we focus on implicit-solvent CG modeling of polymer solutions. The CG model is based on the generalized Langevin equation, where the memory kernel plays the critical role in determining the dynamics in all time scales. Thus, we propose methods for transfer learning of memory kernels. The key ingredient of our methods is Gaussian process regression. By integration with the model order reduction via proper orthogonal decomposition and the active learning technique, the transfer learning can be practically efficient and requires minimum training data. Through two example polymer solution systems, we demonstrate the accuracy and efficiency of the proposed transfer learning methods in the construction of transferable memory kernels. The transferability allows for out-of-sample predictions, even in the extrapolated domain of parameters. Built on the transferable memory kernels, the CG models can reproduce the dynamic properties of polymers in all time scales at different thermodynamic conditions (such as temperature and solvent viscosity) and for different systems with varying concentrations and lengths of polymers.

cs.CE↗

arXiv2021

Data-driven nonintrusive reduced order modeling for dynamical systems with moving boundaries using Gaussian process regression

We present a data-driven nonintrusive model order reduction method for dynamical systems with moving boundaries. The proposed method draws on the proper orthogonal decomposition, Gaussian process regression, and moving least squares interpolation. It combines several attributes that are not simultaneously satisfied in the existing model order reduction methods for dynamical systems with moving boundaries. Specifically, the method requires only snapshot data of state variables at discrete time instances and the parameters that characterize the boundaries, but not further knowledge of the full-order model and the underlying governing equations. The dynamical systems can be generally nonlinear. The movements of boundaries are not limited to prescribed or periodic motions but can be free motions. In addition, we numerically investigate the ability of the reduced order model constructed by the proposed method to forecast the full-order solutions for future times beyond the range of snapshot data. The error analysis for the proposed reduced order modeling and the criteria to determine the furthest forecast time are also provided. Through numerical experiments, we assess the accuracy and efficiency of the proposed method in several benchmark problems. The snapshot data used to construct and validate the reduced order model are from analytical/numerical solutions and experimental measurements.

cs.CE↗

arXiv2021

Advances In Video Compression System Using Deep Neural Network: A Review And Case Studies

Significant advances in video compression system have been made in the past several decades to satisfy the nearly exponential growth of Internet-scale video traffic. From the application perspective, we have identified three major functional blocks including pre-processing, coding, and post-processing, that have been continuously investigated to maximize the end-user quality of experience (QoE) under a limited bit rate budget. Recently, artificial intelligence (AI) powered techniques have shown great potential to further increase the efficiency of the aforementioned functional blocks, both individually and jointly. In this article, we review extensively recent technical advances in video compression system, with an emphasis on deep neural network (DNN)-based approaches; and then present three comprehensive case studies. On pre-processing, we show a switchable texture-based video coding example that leverages DNN-based scene understanding to extract semantic areas for the improvement of subsequent video coder. On coding, we present an end-to-end neural video coding framework that takes advantage of the stacked DNNs to efficiently and compactly code input raw videos via fully data-driven learning. On post-processing, we demonstrate two neural adaptive filters to respectively facilitate the in-loop and post filtering for the enhancement of compressed frames. Finally, a companion website hosting the contents developed in this work can be accessed publicly at https://purdueviper.github.io/dnn-coding/.

eess.IV↗

arXiv2020

Inferring Point Cloud Quality via Graph Similarity

We propose the GraphSIM -- an objective metric to accurately predict the subjective quality of point cloud with superimposed geometry and color impairments. Motivated by the facts that human vision system is more sensitive to the high spatial-frequency components (e.g., contours, edges), and weighs more to the local structural variations rather individual point intensity, we first extract geometric keypoints by resampling the reference point cloud geometry information to form the object skeleton; we then construct local graphs centered at these keypoints for both reference and distorted point clouds, followed by collectively aggregating color gradient moments (e.g., zeroth, first, and second) that are derived between all other points and centered keypoint in the same local graph for significant feature similarity (a.k.a., local significance) measurement; Final similarity index is obtained by pooling the local graph significance across all color channels and by averaging across all graphs. Our GraphSIM is validated using two large and independent point cloud assessment datasets that involve a wide range of impairments (e.g., re-sampling, compression, additive noise), reliably demonstrating the state-of-the-art performance for all distortions with noticeable gains in predicting the subjective mean opinion score (MOS), compared with those point-wise distance-based metrics adopted in standardization reference software. Ablation studies have further shown that GraphSIM is generalized to various scenarios with consistent performance by examining its key modules and parameters.

eess.IV↗

arXiv2020

Multiscale Point Cloud Geometry Compression

Recent years have witnessed the growth of point cloud based applications because of its realistic and fine-grained representation of 3D objects and scenes. However, it is a challenging problem to compress sparse, unstructured, and high-precision 3D points for efficient communication. In this paper, leveraging the sparsity nature of point cloud, we propose a multiscale end-to-end learning framework which hierarchically reconstructs the 3D Point Cloud Geometry (PCG) via progressive re-sampling. The framework is developed on top of a sparse convolution based autoencoder for point cloud compression and reconstruction. For the input PCG which has only the binary occupancy attribute, our framework translates it to a downscaled point cloud at the bottleneck layer which possesses both geometry and associated feature attributes. Then, the geometric occupancy is losslessly compressed using an octree codec and the feature attributes are lossy compressed using a learned probabilistic context model.Compared to state-of-the-art Video-based Point Cloud Compression (V-PCC) and Geometry-based PCC (G-PCC) schemes standardized by the Moving Picture Experts Group (MPEG), our method achieves more than 40% and 70% BD-Rate (Bjontegaard Delta Rate) reduction, respectively. Its encoding runtime is comparable to that of G-PCC, which is only 1.5% of V-PCC.

eess.IV↗

arXiv2020

Neural Video Coding using Multiscale Motion Compensation and Spatiotemporal Context Model

Over the past two decades, traditional block-based video coding has made remarkable progress and spawned a series of well-known standards such as MPEG-4, H.264/AVC and H.265/HEVC. On the other hand, deep neural networks (DNNs) have shown their powerful capacity for visual content understanding, feature extraction and compact representation. Some previous works have explored the learnt video coding algorithms in an end-to-end manner, which show the great potential compared with traditional methods. In this paper, we propose an end-to-end deep neural video coding framework (NVC), which uses variational autoencoders (VAEs) with joint spatial and temporal prior aggregation (PA) to exploit the correlations in intra-frame pixels, inter-frame motions and inter-frame compensation residuals, respectively. Novel features of NVC include: 1) To estimate and compensate motion over a large range of magnitudes, we propose an unsupervised multiscale motion compensation network (MS-MCN) together with a pyramid decoder in the VAE for coding motion features that generates multiscale flow fields, 2) we design a novel adaptive spatiotemporal context model for efficient entropy coding for motion information, 3) we adopt nonlocal attention modules (NLAM) at the bottlenecks of the VAEs for implicit adaptive feature extraction and activation, leveraging its high transformation capacity and unequal weighting with joint global and local information, and 4) we introduce multi-module optimization and a multi-frame training strategy to minimize the temporal error propagation among P-frames. NVC is evaluated for the low-delay causal settings and compared with H.265/HEVC, H.264/AVC and the other learnt video compression methods following the common test conditions, demonstrating consistent gains across all popular test sequences for both PSNR and MS-SSIM distortion metrics.

eess.IV↗

arXiv2020

Dynamic assembly of active colloids: theory and simulation

Because of consuming energy to drive their motion, systems of active colloids are intrinsically out of equilibrium. In the past decade, a variety of intriguing dynamic patterns have been observed in systems of active colloids, and they offer a new platform for studying non-equilibrium physics, in which computer simulation and analytical theory have played an important role. Here we review the recent progress in understanding the dynamic assembly of active colloids by using numerical and analytical tools. We review the progress in understanding the motility induced phase separation in the past decade, followed by the discussion on the effect of shape anisotropy and hydrodynamics on the dynamic assembly of active colloids.

cond-mat.soft↗

arXiv2020

A Dual Camera System for High Spatiotemporal Resolution Video Acquisition

This paper presents a dual camera system for high spatiotemporal resolution (HSTR) video acquisition, where one camera shoots a video with high spatial resolution and low frame rate (HSR-LFR) and another one captures a low spatial resolution and high frame rate (LSR-HFR) video. Our main goal is to combine videos from LSR-HFR and HSR-LFR cameras to create an HSTR video. We propose an end-to-end learning framework, AWnet, mainly consisting of a FlowNet and a FusionNet that learn an adaptive weighting function in pixel domain to combine inputs in a frame recurrent fashion. To improve the reconstruction quality for cameras used in reality, we also introduce noise regularization under the same framework. Our method has demonstrated noticeable performance gains in terms of both objective PSNR measurement in simulation with different publicly available video and light-field datasets and subjective evaluation with real data captured by dual iPhone 7 and Grasshopper3 cameras. Ablation studies are further conducted to investigate and explore various aspects (such as reference structure, camera parallax, exposure time, etc) of our system to fully understand its capability for potential applications.

eess.IV↗

arXiv2020

Object-Based Image Coding: A Learning-Driven Revisit

The Object-Based Image Coding (OBIC) that was extensively studied about two decades ago, promised a vast application perspective for both ultra-low bitrate communication and high-level semantical content understanding, but it had rarely been used due to the inefficient compact representation of object with arbitrary shape. A fundamental issue behind is how to efficiently process the arbitrary-shaped objects at a fine granularity (e.g., feature element or pixel wise). To attack this, we have proposed to apply the element-wise masking and compression by devising an object segmentation network for image layer decomposition, and parallel convolution-based neural image compression networks to process masked foreground objects and background scene separately. All components are optimized in an end-to-end learning framework to intelligently weigh their (e.g., object and background) contributions for visually pleasant reconstruction. We have conducted comprehensive experiments to evaluate the performance on PASCAL VOC dataset at a very low bitrate scenario (e.g., $\lesssim$0.1 bits per pixel - bpp) which have demonstrated noticeable subjective quality improvement compared with JPEG2K, HEVC-based BPG and another learned image compression method. All relevant materials are made publicly accessible at https://njuvision.github.io/Neural-Object-Coding/.

eess.IV↗

arXiv2020

Intelligent Autofocus

We demonstrate that deep learning methods can determine the best focus position from 1-2 image samples, enabling 5-10x faster focus than traditional search-based methods. In contrast with phase detection methods, deep autofocus does not require specialized hardware. In further constrast with conventional methods, which assume a static "best focus," AI methods can generate scene-based focus trajectories that optimize synthesized image quality for dynamic and three dimensional scenes.

eess.IV↗

arXiv2020

Smart Cameras

We review camera architecture in the age of artificial intelligence. Modern cameras use physical components and software to capture, compress and display image data. Over the past 5 years, deep learning solutions have become superior to traditional algorithms for each of these functions. Deep learning enables 10-100x reduction in electrical sensor power per pixel, 10x improvement in depth of field and dynamic range and 10-100x improvement in image pixel count. Deep learning enables multiframe and multiaperture solutions that fundamentally shift the goals of physical camera design. Here we review the state of the art of deep learning in camera operations and consider the impact of AI on the physical design of cameras.

eess.IV↗