arXiv subjects

Mingliang Zhou

Publications and source records attributed to Mingliang Zhou.

31 records · Page 2Linked to original sources

arXiv2024

Unified Image Restoration and Enhancement: Degradation Calibrated Cycle Reconstruction Diffusion Model

Image restoration and enhancement are pivotal for numerous computer vision applications, yet unifying these tasks efficiently remains a significant challenge. Inspired by the iterative refinement capabilities of diffusion models, we propose CycleRDM, a novel framework designed to unify restoration and enhancement tasks while achieving high-quality mapping. Specifically, CycleRDM first learns the mapping relationships among the degraded domain, the rough normal domain, and the normal domain through a two-stage diffusion inference process. Subsequently, we transfer the final calibration process to the wavelet low-frequency domain using discrete wavelet transform, performing fine-grained calibration from a frequency domain perspective by leveraging task-specific frequency spaces. To improve restoration quality, we design a feature gain module for the decomposed wavelet high-frequency domain to eliminate redundant features. Additionally, we employ multimodal textual prompts and Fourier transform to drive stable denoising and reduce randomness during the inference process. After extensive validation, CycleRDM can be effectively generalized to a wide range of image restoration and enhancement tasks while requiring only a small number of training samples to be significantly superior on various benchmarks of reconstruction quality and perceptual quality. The source code will be available at https://github.com/hejh8/CycleRDM.

cs.CV↗

arXiv2024

State Estimation Transformers for Agile Legged Locomotion

We propose a state estimation method that can accurately predict the robot's privileged states to push the limits of quadruped robots in executing advanced skills such as jumping in the wild. In particular, we present the State Estimation Transformers (SET), an architecture that casts the state estimation problem as conditional sequence modeling. SET outputs the robot states that are hard to obtain directly in the real world, such as the body height and velocities, by leveraging a causally masked Transformer. By conditioning an autoregressive model on the robot's past states, our SET model can predict these privileged observations accurately even in highly dynamic locomotions. We evaluate our methods on three tasks -- running jumping, running backflipping, and running sideslipping -- on a low-cost quadruped robot, Cyberdog2. Results show that SET can outperform other methods in estimation accuracy and transferability in the simulation as well as success rates of jumping and triggering a recovery controller in the real world, suggesting the superiority of such a Transformer-based explicit state estimator in highly dynamic locomotion tasks.

cs.RO↗

arXiv2024

Optimizing 4D Lookup Table for Low-light Video Enhancement via Wavelet Priori

Low-light video enhancement is highly demanding in maintaining spatiotemporal color consistency. Therefore, improving the accuracy of color mapping and keeping the latency low is challenging. Based on this, we propose incorporating Wavelet-priori for 4D Lookup Table (WaveLUT), which effectively enhances the color coherence between video frames and the accuracy of color mapping while maintaining low latency. Specifically, we use the wavelet low-frequency domain to construct an optimized lookup prior and achieve an adaptive enhancement effect through a designed Wavelet-prior 4D lookup table. To effectively compensate the a priori loss in the low light region, we further explore a dynamic fusion strategy that adaptively determines the spatial weights based on the correlation between the wavelet lighting prior and the target intensity structure. In addition, during the training phase, we devise a text-driven appearance reconstruction method that dynamically balances brightness and content through multimodal semantics-driven Fourier spectra. Extensive experiments on a wide range of benchmark datasets show that this method effectively enhances the previous method's ability to perceive the color space and achieves metric-favorable and perceptually oriented real-time enhancement while maintaining high efficiency.

cs.CV↗

arXiv2024

Low-light Image Enhancement via CLIP-Fourier Guided Wavelet Diffusion

Low-light image enhancement techniques have significantly progressed, but unstable image quality recovery and unsatisfactory visual perception are still significant challenges. To solve these problems, we propose a novel and robust low-light image enhancement method via CLIP-Fourier Guided Wavelet Diffusion, abbreviated as CFWD. Specifically, CFWD leverages multimodal visual-language information in the frequency domain space created by multiple wavelet transforms to guide the enhancement process. Multi-scale supervision across different modalities facilitates the alignment of image features with semantic features during the wavelet diffusion process, effectively bridging the gap between degraded and normal domains. Moreover, to further promote the effective recovery of the image details, we combine the Fourier transform based on the wavelet transform and construct a Hybrid High Frequency Perception Module (HFPM) with a significant perception of the detailed features. This module avoids the diversity confusion of the wavelet diffusion process by guiding the fine-grained structure recovery of the enhancement results to achieve favourable metric and perceptually oriented enhancement. Extensive quantitative and qualitative experiments on publicly available real-world benchmarks show that our approach outperforms existing state-of-the-art methods, achieving significant progress in image quality and noise suppression. The project code is available at https://github.com/hejh8/CFWD.

cs.CV↗

arXiv2024

EFLNet: Enhancing Feature Learning for Infrared Small Target Detection

Single-frame infrared small target detection is considered to be a challenging task, due to the extreme imbalance between target and background, bounding box regression is extremely sensitive to infrared small target, and target information is easy to lose in the high-level semantic layer. In this article, we propose an enhancing feature learning network (EFLNet) to address these problems. First, we notice that there is an extremely imbalance between the target and the background in the infrared image, which makes the model pay more attention to the background features rather than target features. To address this problem, we propose a new adaptive threshold focal loss (ATFL) function that decouples the target and the background, and utilizes the adaptive mechanism to adjust the loss weight to force the model to allocate more attention to target features. Second, we introduce the normalized Gaussian Wasserstein distance (NWD) to alleviate the difficulty of convergence caused by the extreme sensitivity of the bounding box regression to infrared small target. Finally, we incorporate a dynamic head mechanism into the network to enable adaptive learning of the relative importance of each semantic layer. Experimental results demonstrate our method can achieve better performance in the detection performance of infrared small target compared to the state-of-the-art (SOTA) deep-learning-based methods. The source codes and bounding box annotated datasets are available at https://github.com/YangBo0411/infrared-small-target.

cs.CV↗

arXiv2023

Recent Advances in Rate Control: From Optimisation to Implementation and Beyond

Video coding is a video compression technique that compresses the original video sequence to produce a smaller archive file or reduce the transmission bandwidth under constraints on the visual quality loss. Rate control (RC) plays a critical role in video coding. It can achieve stable stream output in practical applications, especially real-time video applications such as video conferencing or game live streaming. Most RC algorithms either directly or indirectly characterise the relationship between the bit rate (R) and quantisation (Q) and then allocate bits to every coding unit so as to guarantee the global bit rate and video quality level. This paper comprehensively reviews the classic RC technologies used in international video standards of past generations, analyses the mathematical models and implementation mechanisms of various schemes, and compares the performance of recent state-of-the-art RC algorithms. Finally, we discuss future directions and new application areas for RC methods. We hope that this review can help support the development, implementation, and application of RC for new video coding standards.

cs.MM↗

arXiv2023

Effective Community Search on Large Attributed Bipartite Graphs

Community search over bipartite graphs has attracted significant interest recently. In many applications such as user-item bipartite graph in E-commerce, customer-movie bipartite graph in movie rating website, nodes tend to have attributes, while previous community search algorithm on bipartite graphs ignore attributes, which makes the returned results with poor cohesion with respect to their node attributes. In this paper, we study the community search problem on attributed bipartite graphs. Given a query vertex q, we aim to find attributed $\left(α,β\right)$-communities of $G$, where the structure cohesiveness of the community is described by an $\left(α,β\right)$-core model, and the attribute similarity of two groups of nodes in the subgraph is maximized. In order to retrieve attributed communities from bipartite graphs, we first propose a basic algorithm composed of two steps: the generation and verification of candidate keyword sets, and then two improved query algorithms Inc and Dec are proposed. Inc is proposed considering the anti-monotonity property of attributed bipartite graphs, then we adopt different generating method and verifying order of candidate keyword sets and propose the Dec algorithm. After evaluating our solutions on eight large graphs, the experimental results demonstrate that our methods are effective and efficient in querying the attributed communities on bipartite graphs.

cs.SI↗

arXiv2023

Smaller Is Bigger: Rethinking the Embedding Rate of Deep Hiding

Deep hiding, concealing secret information using Deep Neural Networks (DNNs), can significantly increase the embedding rate and improve the efficiency of secret sharing. Existing works mainly force on designing DNNs with higher embedding rates or fancy functionalities. In this paper, we want to answer some fundamental questions: how to increase and what determines the embedding rate of deep hiding. To this end, we first propose a novel Local Deep Hiding (LDH) scheme that significantly increases the embedding rate by hiding large secret images into small local regions of cover images. Our scheme consists of three DNNs: hiding, locating, and revealing. We use the hiding network to convert a secret image in a small imperceptible compact secret code that is embedded into a random local region of a cover image. The locating network assists the revealing process by identifying the position of secret codes in the stego image, while the revealing network recovers all full-size secret images from these identified local regions. Our LDH achieves an extremely high embedding rate, i.e., $16\times24$ bpp and exhibits superior robustness to common image distortions. We also conduct comprehensive experiments to evaluate our scheme under various system settings. We further quantitatively analyze the trade-off between the embedding rate and image quality with different image restoration algorithms.

cs.MM↗

arXiv2022

DeepWSD: Projecting Degradations in Perceptual Space to Wasserstein Distance in Deep Feature Space

Existing deep learning-based full-reference IQA (FR-IQA) models usually predict the image quality in a deterministic way by explicitly comparing the features, gauging how severely distorted an image is by how far the corresponding feature lies from the space of the reference images. Herein, we look at this problem from a different viewpoint and propose to model the quality degradation in perceptual space from a statistical distribution perspective. As such, the quality is measured based upon the Wasserstein distance in the deep feature domain. More specifically, the 1DWasserstein distance at each stage of the pre-trained VGG network is measured, based on which the final quality score is performed. The deep Wasserstein distance (DeepWSD) performed on features from neural networks enjoys better interpretability of the quality contamination caused by various types of distortions and presents an advanced quality prediction capability. Extensive experiments and theoretical analysis show the superiority of the proposed DeepWSD in terms of both quality prediction and optimization.

eess.IV↗

arXiv2020

Regularized Attentive Capsule Network for Overlapped Relation Extraction

Distantly supervised relation extraction has been widely applied in knowledge base construction due to its less requirement of human efforts. However, the automatically established training datasets in distant supervision contain low-quality instances with noisy words and overlapped relations, introducing great challenges to the accurate extraction of relations. To address this problem, we propose a novel Regularized Attentive Capsule Network (RA-CapNet) to better identify highly overlapped relations in each informal sentence. To discover multiple relation features in an instance, we embed multi-head attention into the capsule network as the low-level capsules, where the subtraction of two entities acts as a new form of relation query to select salient features regardless of their positions. To further discriminate overlapped relation features, we devise disagreement regularization to explicitly encourage the diversity among both multiple attention heads and low-level capsules. Extensive experiments conducted on widely used datasets show that our model achieves significant improvements in relation extraction.

cs.CL↗

arXiv2018

Centrality fluctuations in heavy-ion collisions

Volume or centrality fluctuations (CF) is one of the main uncertainties for interpreting the centrality dependence of many experimental observables. The CF is constrained by centrality selection based on particle multiplicity in a reference subevent, and contributes to observables measured in another subevent. Using a Glauber-based independent source model, we study the influence of CF on several distributions of multiplicity $N$ and eccentricities $ε_n$: $p(N)$, $p(ε_n)$, $p(ε_n,ε_m)$ and $p(N,ε_n)$, where the effects of CF is quantified using multi-particle cumulants of these distributions. In mid-central collisions, a general relation is established between the multiplicity fluctuation and resulting CF in the reference subevent. In ultra-central collisions, where distribution of particle production sources is strongly distorted, we find these cumulants exhibit rich sign-change patterns, due to observable-dependent non-Gaussianity in the underlying distributions. The details of sign-change pattern change with the size of the collision systems. Simultaneous comparison of these different types cumulants between model prediction and experimental data can be used to constrain the CF and particle production mechanism in heavy-ion collisions. Since the concept of centrality and CF are expected to fluctuate in the longitudinal direction within a single event, we propose to use pseudorapidity-separated subevent cumulant method to explore the nature of intra-event fluctuations of centrality and collective dynamics. The subevent method can be applied for any bulk observable that is sensitive to centrality, and has the potential to separate different mechanisms for multiplicity and flow fluctuations happening at different time scales. The forward detector upgrades at RHIC and LHC will greatly enhance such studies in the future.

nucl-th↗

arXiv2017

Revealing long-range multi-particle collectivity in small collision systems via subevent cumulants

Multi-particle azimuthal cumulants, often used to study collective flow in high-energy heavy-ion collisions, have recently been applied in small collision systems such as $pp$ and $p$+A to extract the second-order azimuthal harmonic flow $v_2$. Recent observation of four-, six- and eight-particle cumulants with "correct sign" $c_2\{4\}<0, c_2\{6\}>0, c_2\{8\}<0$ and approximate equality of the inferred single-particle harmonic flow, $v_2\{4\}\approx v_2\{6\}\approx v_2\{8\}$, have been used as strong evidence for a collective emission of all soft particles produced in the collisions. We show that these relations in principle could be violated due to the non-Gaussianity in the event-by-event fluctuation of flow and/or non-flow. Furthermore, we show, using $pp$ events generated with the PYTHIA model, that $c_2\{2k\}$ obtained with standard cumulant method are dominated by non-flow from dijets. An alternative cumulant method based on two or more $η$-separated subevents is proposed to suppress the dijet contribution. The new method is shown to be able to recover a flow signal as low as 4\% imposed on the PYTHIA events, independently of how the event activity class is defined. Therefore the subevent cumulant method offers a more robust way of studying collectivity based on the existence of long-range azimuthal correlations between multiple distinct $η$ ranges. The prospect of using the subevent cumulants to study collective flow in A+A collisions, in particular its longitudinal dynamics, is discussed.

nucl-th↗

arXiv2016

Forward-backward multiplicity fluctuation and longitudinal harmonics in high-energy nuclear collisions

An analysis method is proposed to study the forward-backward (FB) multiplicity fluctuation in high-energy nuclear collisions, built on the earlier work of Bzdak and Teaney. The method allows the decomposition of the centrality dependence of average multiplicity from the dynamical event-by-event (EbyE) fluctuation of multiplicity in pseudorapidity. Application of the method to AMPT and HIJING models shows that the long-range component of the FB correlation is captured by a few longitudinal harmonics, with the first component driven by the asymmetry in the number of participating nucleons in the two colliding nuclei. The higher-order longitudinal harmonics are found to be strongly damped in AMPT compare to HIJING, due to weaker short-range correlations as well as the final-state effects present in the AMPT model. Two-particle pseudorapidity correlation reveals interesting charge-dependent short-range structures that are absent in HIJING model. The proposed method opens an avenue to elucidate the particle production mechanism and early time dynamics in heavy-ion collisions. Future analysis directions and prospects of using the pseudorapidity correlation function to understand the centrality bias in $p$+$p$, $p$+A and A+A collisions are discussed.

nucl-th↗