Search

AI-generated video has revolutionized short video production, filmmaking, and personalized media, making video local editing an essential tool. However, this progress also blurs the line between reality and fiction, posing challenges in multimedia forensics. To solve this urgent issue, V2A-Mark is proposed to address the limitations of current video tampering forensics, such as poor generalizability, singular function, and single modality focus. Combining the fragility of video-into-video steganography with deep robust watermarking, our method can embed invisible visual-audio localization watermarks and copyright watermarks into the original video frames and audio, enabling precise manipulation localization and copyright protection. We also design a temporal alignment and fusion module and degradation prompt learning to enhance the localization accuracy and decoding robustness. Meanwhile, we introduce a sample-level audio localization method and a cross-modal copyright extraction mechanism to couple the information of audio and video frames. The effectiveness of V2A-Mark has been verified on a visual-audio tampering dataset, emphasizing its superiority in localization precision and copyright accuracy, crucial for the sustainable development of video editing in the AIGC video era.

With the advent of virtual reality technology, omnidirectional image (ODI) rescaling techniques are increasingly embraced for reducing transmitted and stored file sizes while preserving high image quality. Despite this progress, current ODI rescaling methods predominantly focus on enhancing the quality of images in equirectangular projection (ERP) format, which overlooks the fact that the content viewed on head mounted displays (HMDs) is actually a rendered viewport instead of an ERP image. In this work, we emphasize that focusing solely on ERP quality results in inferior viewport visual experiences for users. Thus, we propose ResVR, which is the first comprehensive framework for the joint Rescaling and Viewport Rendering of ODIs. ResVR allows obtaining LR ERP images for transmission while rendering high-quality viewports for users to watch on HMDs. In our ResVR, a novel discrete pixel sampling strategy is developed to tackle the complex mapping between the viewport and ERP, enabling end-to-end training of ResVR pipeline. Furthermore, a spherical pixel shape representation technique is innovatively derived from spherical differentiation to significantly improve the visual quality of rendered viewports. Extensive experiments demonstrate that our ResVR outperforms existing methods in viewport rendering tasks across different fields of view, resolutions, and view directions while keeping a low transmission overhead.

Omnidirectional images (ODIs) are commonly used in real-world visual tasks, and high-resolution ODIs help improve the performance of related visual tasks. Most existing super-resolution methods for ODIs use end-to-end learning strategies, resulting in inferior realness of generated images and a lack of effective out-of-domain generalization capabilities in training methods. Image generation methods represented by diffusion model provide strong priors for visual tasks and have been proven to be effectively applied to image restoration tasks. Leveraging the image priors of the Stable Diffusion (SD) model, we achieve omnidirectional image super-resolution with both fidelity and realness, dubbed as OmniSSR. Firstly, we transform the equirectangular projection (ERP) images into tangent projection (TP) images, whose distribution approximates the planar image domain. Then, we use SD to iteratively sample initial high-resolution results. At each denoising iteration, we further correct and update the initial results using the proposed Octadecaplex Tangent Information Interaction (OTII) and Gradient Decomposition (GD) technique to ensure better consistency. Finally, the TP images are transformed back to obtain the final high-resolution results. Our method is zero-shot, requiring no training or fine-tuning. Experiments of our method on two benchmark datasets demonstrate the effectiveness of our proposed method.

While deep neural networks (NN) significantly advance image compressed sensing (CS) by improving reconstruction quality, the necessity of training current CS NNs from scratch constrains their effectiveness and hampers rapid deployment. Although recent methods utilize pre-trained diffusion models for image reconstruction, they struggle with slow inference and restricted adaptability to CS. To tackle these challenges, this paper proposes Invertible Diffusion Models (IDM), a novel efficient, end-to-end diffusion-based CS method. IDM repurposes a large-scale diffusion sampling process as a reconstruction model, and finetunes it end-to-end to recover original images directly from CS measurements, moving beyond the traditional paradigm of one-step noise estimation learning. To enable such memory-intensive end-to-end finetuning, we propose a novel two-level invertible design to transform both (1) the multi-step sampling process and (2) the noise estimation U-Net in each step into invertible networks. As a result, most intermediate features are cleared during training to reduce up to 93.8% GPU memory. In addition, we develop a set of lightweight modules to inject measurements into noise estimator to further facilitate reconstruction. Experiments demonstrate that IDM outperforms existing state-of-the-art CS networks by up to 2.64dB in PSNR. Compared to the recent diffusion model-based approach DDNM, our IDM achieves up to 10.09dB PSNR gain and 14.54 times faster inference.

Panorama video recently attracts more interest in both study and application, courtesy of its immersive experience. Due to the expensive cost of capturing 360-degree panoramic videos, generating desirable panorama videos by prompts is urgently required. Lately, the emerging text-to-video (T2V) diffusion methods demonstrate notable effectiveness in standard video generation. However, due to the significant gap in content and motion patterns between panoramic and standard videos, these methods encounter challenges in yielding satisfactory 360-degree panoramic videos. In this paper, we propose a pipeline named 360-Degree Video Diffusion model (360DVD) for generating 360-degree panoramic videos based on the given prompts and motion conditions. Specifically, we introduce a lightweight 360-Adapter accompanied by 360 Enhancement Techniques to transform pre-trained T2V models for panorama video generation. We further propose a new panorama dataset named WEB360 consisting of panoramic video-text pairs for training 360DVD, addressing the absence of captioned panoramic video datasets. Extensive experiments demonstrate the superiority and effectiveness of 360DVD for panorama video generation. Our project page is at https://akaneqwq.github.io/360DVD/., Comment: arXiv admin note: text overlap with arXiv:2307.04725 by other authors

In the era where AI-generated content (AIGC) models can produce stunning and lifelike images, the lingering shadow of unauthorized reproductions and malicious tampering poses imminent threats to copyright integrity and information security. Current image watermarking methods, while widely accepted for safeguarding visual content, can only protect copyright and ensure traceability. They fall short in localizing increasingly realistic image tampering, potentially leading to trust crises, privacy violations, and legal disputes. To solve this challenge, we propose an innovative proactive forensics framework EditGuard, to unify copyright protection and tamper-agnostic localization, especially for AIGC-based editing methods. It can offer a meticulous embedding of imperceptible watermarks and precise decoding of tampered areas and copyright information. Leveraging our observed fragility and locality of image-into-image steganography, the realization of EditGuard can be converted into a united image-bit steganography issue, thus completely decoupling the training process from the tampering types. Extensive experiments demonstrate that our EditGuard balances the tamper localization accuracy, copyright recovery precision, and generalizability to various AIGC-based tampering methods, especially for image forgery that is difficult for the naked eye to detect. The project page is available at https://xuanyuzhang21.github.io/project/editguard/., Comment: AIGC image watermarking, tamper localization and copyright protection

Real-world data is extremely imbalanced and presents a long-tailed distribution, resulting in models that are biased towards classes with sufficient samples and perform poorly on rare classes. Recent methods propose to rebalance classes but they undertake the seesaw dilemma (what is increasing performance on tail classes may decrease that of head classes, and vice versa). In this paper, we argue that the seesaw dilemma is derived from gradient imbalance of different classes, in which gradients of inappropriate classes are set to important for updating, thus are prone to overcompensation or undercompensation on tail classes. To achieve ideal compensation, we formulate the long-tailed recognition as an multi-objective optimization problem, which fairly respects the contributions of head and tail classes simultaneously. For efficiency, we propose a Gradient-Balancing Grouping (GBG) strategy to gather the classes with similar gradient directions, thus approximately make every update under a Pareto descent direction. Our GBG method drives classes with similar gradient directions to form more representative gradient and provide ideal compensation to the tail classes. Moreover, We conduct extensive experiments on commonly used benchmarks in long-tailed learning and demonstrate the superiority of our method over existing SOTA methods., Comment: In submission

Multi-stage strategies are frequently employed in image restoration tasks. While transformer-based methods have exhibited high efficiency in single-image super-resolution tasks, they have not yet shown significant advantages over CNN-based methods in stereo super-resolution tasks. This can be attributed to two key factors: first, current single-image super-resolution transformers are unable to leverage the complementary stereo information during the process; second, the performance of transformers is typically reliant on sufficient data, which is absent in common stereo-image super-resolution algorithms. To address these issues, we propose a Hybrid Transformer and CNN Attention Network (HTCAN), which utilizes a transformer-based network for single-image enhancement and a CNN-based network for stereo information fusion. Furthermore, we employ a multi-patch training strategy and larger window sizes to activate more input pixels for super-resolution. We also revisit other advanced techniques, such as data augmentation, data ensemble, and model ensemble to reduce overfitting and data bias. Finally, our approach achieved a score of 23.90dB and emerged as the winner in Track 1 of the NTIRE 2023 Stereo Image Super-Resolution Challenge., Comment: 10 pages, 3 figures, accepted by CVPR workshop 2023

360{\deg} omnidirectional images have gained research attention due to their immersive and interactive experience, particularly in AR/VR applications. However, they suffer from lower angular resolution due to being captured by fisheye lenses with the same sensor size for capturing planar images. To solve the above issues, we propose a two-stage framework for 360{\deg} omnidirectional image superresolution. The first stage employs two branches: model A, which incorporates omnidirectional position-aware deformable blocks (OPDB) and Fourier upsampling, and model B, which adds a spatial frequency fusion module (SFF) to model A. Model A aims to enhance the feature extraction ability of 360{\deg} image positional information, while Model B further focuses on the high-frequency information of 360{\deg} images. The second stage performs same-resolution enhancement based on the structure of model A with a pixel unshuffle operation. In addition, we collected data from YouTube to improve the fitting ability of the transformer, and created pseudo low-resolution images using a degradation network. Our proposed method achieves superior performance and wins the NTIRE 2023 challenge of 360{\deg} omnidirectional image super-resolution., Comment: Accepted to CVPRW 2023

The electronic transport properties of Janus monolayer black arsenic phosphorus (b-AsP) nanoribbons have been investigated utilizing the tight-binding approach. The dependence of electronic structure on edge structures is systematically investigated. (1,3)nb and (3,1)nb b-AsP nanoribbons exhibit flatter edge bands than zigzag and armchair counterparts. The edge band of the armchair ones show double degeneracy. Further, the calculated results show the band gap of the zigzag ribbon with different boundary morphology is widely tunable by transverse electric field. A critical electric field can fully close the gap and induce a phase transition from a semiconductor into a conductor. Our work suggests dynamically tunable bandgap in Janus b-AsP nanoribbons and reveals the potential of Janus b-AsP for transmission devices., Comment: 15 pages, 9 figures

In this work, we investigate the electronic and optical properties of the b-AsP quantum dots(QDs) and quantum rings(QRs) with different edge types in the presence of an in-plane electric field and a perpendicular magnetic field utilizing the tight-binding method. Our calculations show that the electronic and optical properties of edge states largely depend on the edge types. By adjusting the intensity of the electric field or magnetic field, the probability density can be effectively tuned. In particular, under an in-plane electric field, novel edge states emerge in rhombus QDs and QRs with {\delta} atoms, and transitions happen between these novel edge states and bulk states in the conduction band region, which is illustrated by distinct and evenly separated peaks in the optical absorption spectra. This work might exploit a potential system to implement quantum state transfer among two-dimensional material quantum dots., Comment: 12 pages, 12 figures

Mapping optimization algorithms into neural networks, deep unfolding networks (DUNs) have achieved impressive success in compressive sensing (CS). From the perspective of optimization, DUNs inherit a well-defined and interpretable structure from iterative steps. However, from the viewpoint of neural network design, most existing DUNs are inherently established based on traditional image-domain unfolding, which takes one-channel images as inputs and outputs between adjacent stages, resulting in insufficient information transmission capability and inevitable loss of the image details. In this paper, to break the above bottleneck, we first propose a generalized dual-domain optimization framework, which is general for inverse imaging and integrates the merits of both (1) image-domain and (2) convolutional-coding-domain priors to constrain the feasible region in the solution space. By unfolding the proposed framework into deep neural networks, we further design a novel Dual-Domain Deep Convolutional Coding Network (D3C2-Net) for CS imaging with the capability of transmitting high-throughput feature-level image representation through all the unfolded stages. Experiments on natural and MR images demonstrate that our D3C2-Net achieves higher performance and better accuracy-complexity trade-offs than other state-of-the-arts.

The distributed matrix multiplication problem with an unknown number of stragglers is considered, where the goal is to efficiently and flexibly obtain the product of two massive matrices by distributing the computation across N servers. There are up to N - R stragglers but the exact number is not known a priori. Motivated by reducing the computation load of each server, a flexible solution is proposed to fully utilize the computation capability of available servers. The computing task for each server is separated into several subtasks, constructed based on Entangled Polynomial codes by Yu et al. The final results can be obtained from either a larger number of servers with a smaller amount of computation completed per server or a smaller number of servers with a larger amount of computation completed per server. The required finite field size of the proposed solution is less than 2N. Moreover, the optimal design parameters such as the partitioning of the input matrices is discussed. Our constructions can also be generalized to other settings such as batch distributed matrix multiplication and secure distributed matrix multiplication.

This paper presents flexible storage codes, a class of error-correcting codes that can recover information from a flexible number of storage nodes. As a result, one can make a better use of the available storage nodes in the presence of unpredictable node failures and reduce the data access latency. Let us assume a storage system encodes $k\ell$ information symbols over a finite field $\mathbb{F}$ into $n$ nodes, each of size $\ell$ symbols. The code is parameterized by a set of tuples $\{(R_j,k_j,\ell_j): 1 \le j \le a\}$, satisfying $k_1\ell_1=k_2\ell_2=...=k_a\ell_a$ and $k_1>k_2>...>k_a = k, \ell_a=\ell$, such that the information symbols can be reconstructed from any $R_j$ nodes, each node accessing $\ell_j$ symbols. In other words, the code allows a flexible number of nodes for decoding to accommodate the variance in the data access time of the nodes. Code constructions are presented for different storage scenarios, including LRC (locally recoverable) codes, PMDS (partial MDS) codes, and MSR (minimum storage regenerating) codes. We analyze the latency of accessing information and perform simulations on Amazon clusters to show the efficiency of presented codes.

We theoretically investigate the one-color injection currents and shift currents in zigzag graphene nanoribbons with applying a static electric field across the ribbon, which breaks the inversion symmetry to generate nonzero second order optical responses by dipole interaction. These two types of currents can be separately excited by specific light polarization, circularly polarized lights for injection currents and linearly polarized lights for shift currents. Based on a tight binding model formed by carbon 2p$_z$ orbitals, we numerically calculate the spectra of injection coefficients and shift conductivities, as well as their dependence on the static field strength and ribbon width. The spectra show many peaks associated with the optical transition between different subbands, and the positions and amplitudes of these peaks can be effectively controlled by the static electric field. By constructing a simple two band model, the static electric fields are found to modify the edge states in a nonperturbative way, and their associated optical transitions dominate the current generation at low photon energies. For typical parameters, such as a static field 10$^6$ V/m and light intensity 0.1 GW/cm$^2$, the magnitude of the injection and shift currents for a ribbon with width 5 nm can be as large as the order of 1 $\mu$A. Our results provide a physical basis for realizing passive optoelectronic devices based on graphene nanoribbons.