Diffusion model-based networks have been widely applied in the field of image generation and have gradually demonstrated a strong potential in image colorization tasks. However, despite the emergence of various colorization diffusion models, two major challenges remain: (1) the lack of effective control over the colorization process and (2) the prevalent issue of color bleeding. Integrating suitable conditional control can effectively alleviate these challenges. To this end, we propose a unified multi-modal diffusion model that harnesses diverse modality information to achieve flexible and high-quality colorization. Specifically, we introduce a Stroke-Adapter that extracts and integrates stroke prompt, enhancing user control over color distribution. Additionally, we design an Edge-Guided Attention mechanism to effectively inject edge information into the colorization process, significantly reducing color bleeding artifacts. Extensive comparative experiments demonstrate that our method outperforms state-of-the-art image colorization approaches in both qualitative and quantitative evaluations, achieving superior colorization results with enhanced controllability.

Recently, the mutual combination of Neural Radiance Fields (NeRF) and Segment Anything Model (SAM) in 3D semantic segmentation has achieved impressive results. However, they face the challenge of accurately and consistently segmenting objects in complex scenarios. To address this issue, we introduce the Enhancing Segmentation in NeRF with CLIP(ES-NeRF), which aims to improve the segmentation quality through feature fusion with the help of CLIP’s powerful semantic comprehension. Specifically, we propose a CLIP2SAM module, which utilizes the image-text features extracted by CLIP for cross-modal multi-scale interactions to obtain the semantic features of CLIP on rough segmentation. These features will then be aligned with those extracted by SAM to achieve feature fusion and complete segmentation. Finally, NeRF is employed to aggregate masks from disparate viewpoints, thereby attaining high-quality 3D segmentation. The efficacy of our method is substantiated by a multitude of experimental results, demonstrating its superiority over existing techniques.

Generative artificial intelligence (AI) is versatile for various applications, but security and privacy concerns with third-party AI vendors hinder its broader adoption in sensitive scenarios. Hence, it is essential for users to validate the AI trustworthiness and ensure the security of data boundaries. In this paper, we present a dye testing system named Dye4AI, which injects crafted trigger data into human-AI dialogue and observes AI responses towards specific prompts to diagnose data flow in AI model evolution. Our dye testing procedure contains 3 stages: trigger generation, trigger insertion, and trigger retrieval. First, to retain both uniqueness and stealthiness, we design a new trigger that transforms a pseudo-random number to a intelligible format. Second, with a custom-designed three-step conversation strategy, we insert each trigger item into dialogue and confirm the model memorizes the new trigger knowledge in the current session. Finally, we routinely try to recover triggers with specific prompts in new sessions, as triggers can present in new sessions only if AI vendors leverage user data for model fine-tuning. Extensive experiments on six LLMs demonstrate our dye testing scheme is effective in ensuring the data boundary, even for models with various architectures and parameter sizes. Also, larger and premier models tend to be more suitable for Dye4AI, e.g., trigger can be retrieved in OpenLLaMa-13B even with only 2 insertions per trigger item. Moreover, we analyze the prompt selection in dye testing, providing insights for future testing systems on generative AI services.

Anomaly detection identifies instances that deviate the distribution of the normal class. Recently, the diffusion models have shown great promise. Our research revealed that by training the diffusion model solely on normal data, it is able to transform both normal and anomalous samples into normal images. Employing this discovery, we propose ODD (One-Class Anomaly Detection via the Diffusion model), which consists of: a diffusion model to convert both normal and anomalous data into normal data, and a similarity network enhanced with outlier exposure to measure the semantic distance between the input and output of the diffusion model. If the score is low, the input is considered as an anomaly instance. The ODD is evaluated on a variety of datasets. Both qualitative and quantitative results demonstrate that our method outperforms existing state-of-the-art techniques.

Video colorization aims to add color to black-and-white films. However, propagating color information to the whole video clip accurately is a challenging task. In this paper, we propose Flow-Guided Transformer for Video Colorization (FGTVC), consisting of a Global Motion Aggregation (GMA) module, Residual modules, Flow-Guided Attention blocks (FGAB) based on encoder and decoder, to exploit the information from the neighbor patch with high similarity for each video patch colorization. Specifically, we employ Transformer to capture the long-distance dependencies between frames and learn non-local self-similarity in the frame. To overcome the shortcomings of previous optical flow-based methods, FGAB enjoys the guidance of optical flow to sample elements from spatio-temporal adjacent frames when calculating self-attention. Experiments show that the proposed FGTVC has an outstanding performance than the state-of-the-art methods. In addition, comprehensive findings demonstrate the superiority of our framework in real-world video colorization tasks.

Cloud providers are in a position to greatly improve the trust clients have in network services: IaaS platforms can isolate services so they cannot leak data, and can help verify that they are securely deployed. We describe a new system called CQSTR that allows clients to verify a service's security properties. CQSTR provides a new cloud container abstraction similar to Linux containers but for VM clusters within IaaS clouds. Cloud containers enforce constraints on what software can run, and control where and how much data can be communicated across service boundaries. With CQSTR, IaaS providers can make assertions about the security properties of a service running in the cloud.

Virtualized environments are widely thought to cause problems for software-based random number generators (RNGs), due to use of virtual machine (VM) snapshots as well as fewer and believed-to-be lower quality entropy sources. Despite this, we are unaware of any published analysis of the security of critical RNGs when running in VMs. We fill this gap, using measurements of Linux's RNG systems (without the aid of hardware RNGs, the most common use case today) on Xen, VMware, and Amazon EC2. Despite CPU cycle counters providing a significant source of entropy, various deficiencies in the design of the Linux RNG makes its first output vulnerable during VM boots and, more critically, makes it suffer from catastrophic reset vulnerabilities. We show cases in which the RNG will output the exact same sequence of bits each time it is resumed from the same snapshot. This can compromise, for example, cryptographic secrets generated after resumption. We explore legacy-compatible countermeasures, as well as a clean-slate solution. The latter is a new RNG called Whirlwind that provides a simpler, more-secure solution for providing system randomness.

The cloud has become a promising alternative to traditional HPC centers or in-house clusters. This new environment highlights the I/O bottleneck problem, typically with top-of-the-line compute instances but sub-par communication and I/O facilities. It has been observed that changing cloud I/O system configurations leads to significant variation in the performance and cost efficiency of I/O intensive HPC applications. However, storage system configuration is tedious and error-prone to do manually, even for experts.

Targets of interest in video acquired from imaging infrared sensors often exhibit profound appearance variations due to a variety of factors, including complex target maneuvers, ego-motion of the sensor platform, background clutter, etc., making it difficult to maintain a reliable detection process and track lock over extended time periods. Two key issues in overcoming this problem are how to represent the target and how to learn its appearance online. In this paper, we adopt a recent appearance model that estimates the pixel intensity histograms as well as the distribution of local standard deviations in both the foreground and background regions for robust target representation. Appearance learning is then cast as an adaptive Kalman filtering problem where the process and measurement noise variances are both unknown. We formulate this problem using both covariance matching and, for the first time in a visual tracking application, the recent autocovariance least-squares (ALS) method. Although convergence of the ALS algorithm is guaranteed only for the case of globally wide sense stationary process and measurement noises, we demonstrate for the first time that the technique can often be applied with great effectiveness under the much weaker assumption of piecewise stationarity. The performance advantages of the ALS method relative to the classical covariance matching are illustrated by means of simulated stationary and nonstationary systems. Against real data, our results show that the ALS-based algorithm outperforms the covariance matching as well as the traditional histogram similarity-based methods, achieving sub-pixel tracking accuracy against the well-known AMCOM closure sequences and the recent SENSIAC automatic target recognition dataset.

Belief propagation (BP) is a powerful algorithm to decode low-density parity check (LDPC) codes over additive white Gaussian noise (AWGN) channels. However, the traditional BP algorithm cannot adapt efficiently to the statistical change of SNR in an AWGN channel. This paper proposes an adaptive scheme that incorporates a particle filtering (PF) algorithm into the BP based LDPC decoding process. The proposed scheme is capable to perform online estimation of time-varying SNR at the bit-level and enhance the BP decoding performance simultaneously.

We present a new particle filter (PF) algorithm, which uses a mathematical tool known as Galerkin's projection method to generate the proposal distribution. By definition, Galerkin's method is a numerical approach to approximate the solution of a partial differential equation. By leveraging this method with L2theory and the FFT, this new proposal is fundamentally different to various local linearization or Kalman filter based proposals. We apply this algorithm to a bearings-only tracking problem. As shown in the theory and indicated by our simulations, this proposal renders more support from the true posterior distribution, thereby significantly enhances the estimation accuracy compared to standard bootstrap filters. In addition, because of this improved proposal distribution, the new particle filter can achieve a given level of performance with less sample size.

The adaptive Volterra filter has been successfully applied in nonlinear acoustic echo cancellation (AEC) systems and nonlinear line echo cancellation systems, but its applications are limited by its required computational complexity and slow convergence rate, especially for systems with long memory length. In this paper, by leveraging a multi-channel configuration of the Volterra filter and the sampling theory for nonlinear systems, we extend linear sub-band delay-less adaptive filter techniques to develop an efficient sub-band implementation of the adaptive Volterra filter. The developed sub-band configuration of the adaptive Volterra filter can greatly improve the convergence rate and reduce the computational complexity of nonlinear echo cancellers, which is shown by analyses and simulations.

In this paper, we present a new particle filtering (PF) algorithm for visual target tracking where Galerkin's projection method is used to generate the proposal distribution. Galerkin's method is a numerical approach to approximate the solution of a partial differential equation (PDE). By leveraging this method in concert with L2theory and the FFT, we obtain a new proposal which directly approximates the true state posterior distribution and is fundamentally different from various local linearizations or Kalman filter-based proposals. We apply this improved PF algorithm to track a human head in a video sequence. As predicted by theory and demonstrated by our experimental results, this new algorithm is highly effective for tracking targets which exhibit complex kinematics. The new proposal distribution given here captures the high probability area in the state space, thereby gleaning increased support from the true posterior distribution.

Cyber attacks against networked computers have become relentless in recent years. The most common attack method is to exploit memory corruption vulnerabilities such as buffer overflow and format string bugs. This paper presents a technique to automatically identify both known and unknown memory corruption vulnerabilities. Based on the observation that a randomized program usually crashes upon a memory corruption attack, this technique uses the crash as a trigger to initiate an automatic diagnosis algorithm. The output of the diagnosis includes the instruction that is tricked to corrupt data, the call stack at the time of corruption, and the propagation history of corrupted data. These results provide useful information in fixing the vulnerabilities. Moreover, the diagnosis process also generates a signature of the attack using data/address values embedded in the malicious input message, and is used to block future attacks. Such a signature is further associated with the program execution state to reduce false positives without decreasing the detection rate. The proposed techniques enable the development of a decentralized self-diagnosing and self-protecting defense mechanism for networked computers. We report the implementation experience and experimental evaluation of a prototype system on Linux.

In this paper we propose a novel image restoration method that effectively combines a particle filter with wavelet shrinkage to achieve robust performance against inhomogeneous noise mixtures. Specifically, the particle filter acts to suppress outlier-rich components of the noise while, in a subsequent step, the wavelet domain shrinkage attenuates any remaining, less heavily tailed noise components. We present late breaking preliminary examples demonstrating excellent rejection of salt-and-pepper like Cauchy noise mixed with additive white Gaussian noise (AWGN). Although limited in scope, these preliminary results suggest that the combination of particle filters with more traditional restoration techniques is a powerful approach that can provide a new dimension of flexibility for addressing noise mixtures involving difficult nonlinear and non-Gaussian components.

Sequential importance sampling (SIS), also known as particle filtering, has drawn increasing attention recently due to its superior performance in nonlinear and non-Gaussian dynamic problems. In the SIS framework, estimation accuracy depends strongly on the choice of proposal distribution. In this paper we propose a novel SIS algorithm called PF-SP-PEKF that is based on a state partition technique and a parallel bank of extended Kalman filters designed to improve the accuracy of the proposal distribution. Our results show that this new approach yields a significantly improved estimate of the state, enabling the new particle filter to effectively track human subjects in a video sequence where the standard condensation filter fails to maintain track lock. Moreover, because of the improved proposal distribution, the new filter can achieve a given level of performance using fewer particles than its conventional SIS counterparts.

This paper presents techniques to integrate and reason about complementary intrusion evidence such as alerts generated by intrusion detection systems (IDSs) and reports by system monitoring or vulnerability scanning tools. To facilitate the modeling of intrusion evidence, this paper classifies intrusion evidence into either event-based evidence or state-based evidence. Event-based evidence refers to observations (or detections) of intrusive actions (e.g., IDS alerts), while state-based evidence refers to observations of the effects of intrusions on system states. Based on the interdependency between event-based and state-based evidence, this paper develops techniques to automatically integrate complementary evidence into Bayesian networks, and reason about uncertain or unknown intrusion evidence based on verified evidence. The experimental results in this paper demonstrate the potential of the proposed techniques. In particular, additional observations by system monitoring or vulnerability scanning tools can potentially reduce the false alert rate and increase the confidence in alerts corresponding to successful attacks.