Search

Learned visual compression is an important and active task in multimedia. Existing approaches have explored various CNN- and Transformer-based designs to model content distribution and eliminate redundancy, where balancing efficacy (i.e., rate-distortion trade-off) and efficiency remains a challenge. Recently, state-space models (SSMs) have shown promise due to their long-range modeling capacity and efficiency. Inspired by this, we take the first step to explore SSMs for visual compression. We introduce MambaVC, a simple, strong and efficient compression network based on SSM. MambaVC develops a visual state space (VSS) block with a 2D selective scanning (2DSS) module as the nonlinear activation function after each downsampling, which helps to capture informative global contexts and enhances compression. On compression benchmark datasets, MambaVC achieves superior rate-distortion performance with lower computational and memory overheads. Specifically, it outperforms CNN and Transformer variants by 9.3% and 15.6% on Kodak, respectively, while reducing computation by 42% and 24%, and saving 12% and 71% of memory. MambaVC shows even greater improvements with high-resolution images, highlighting its potential and scalability in real-world applications. We also provide a comprehensive comparison of different network designs, underscoring MambaVC's advantages. Code is available at https://github.com/QinSY123/2024-MambaVC., Comment: 17pages,15 figures

In this paper, a heterogeneous leader-followers multiagent system is studied under simultaneous time-varying communication faults and actuator faults. First, the state of the leader is modelled as the closed-loop reference model where the states of the direct-connected followers are fed to the leader to improve the leader-followers tracking capability. An event-triggered communication mechanism is then designed for the agent information sharing among its neighbors so as to reduce the communication burden. Considering the time-varying communication link failure, a new distributed event-triggered observer is designed for each follower to estimate the whole system states so as to reduce the state error, whereas an adaptive distributed event-triggered estimator is further designed for the nondirect connected followers to estimate the coefficient matrix of the leader system. Further, an estimator is designed for the actuator fault estimation to reduce their impact on the system consistency. Hence, an adaptive event-triggered control strategy is proposed to ensure the consistency of the leader-follower system under the time-varying communication link faults and actuator faults. It is also shown that Zeno behavior is excluded for each agent and the effectiveness of the proposed adaptive event-triggered control strategy is verified on the heterogeneous multi-agent system.

In this paper, we propose a progressive learning paradigm for transformer-based variable-rate image compression. Our approach covers a wide range of compression rates with the assistance of the Layer-adaptive Prompt Module (LPM). Inspired by visual prompt tuning, we use LPM to extract prompts for input images and hidden features at the encoder side and decoder side, respectively, which are fed as additional information into the Swin Transformer layer of a pre-trained transformer-based image compression model to affect the allocation of attention region and the bits, which in turn changes the target compression ratio of the model. To ensure the network is more lightweight, we involves the integration of prompt networks with less convolutional layers. Exhaustive experiments show that compared to methods based on multiple models, which are optimized separately for different target rates, the proposed method arrives at the same performance with 80% savings in parameter storage and 90% savings in datasets. Meanwhile, our model outperforms all current variable bitrate image methods in terms of rate-distortion performance and approaches the state-of-the-art fixed bitrate image compression methods trained from scratch.

There were 27 novel coronavirus pneumonia cases found in Wuhan, China in December 2019, named as 2019-nCoV temporarily and COVID-19 formally by WHO on 11 February, 2020. In December 2019 and January 2020, COVID-19 has spread in large scale among the population, which brought terrible disaster to the life and property of the Chinese people. In this paper, we will first analyze the feature and pattern of the virus transmission, and discuss the key impact factors and uncontrollable factors of epidemic transmission based on public data. Then the virus transmission can be modelled and used for the inflexion and extinction period of epidemic development so as to provide theoretical support for the Chinese government in the decision-making of epidemic prevention and recovery of economic production. Further, this paper demonstrates the effectiveness of the prevention methods taken by the Chinese government such as multi-level administrative region isolation. It is of great importance and practical significance for the world to deal with public health emergencies., Comment: 7 pages, 18 figures

We explore the use of a knowledge graphs, that capture general or commonsense knowledge, to augment the information extracted from images by the state-of-the-art methods for image captioning. The results of our experiments, on several benchmark data sets such as MS COCO, as measured by CIDEr-D, a performance metric for image captioning, show that the variants of the state-of-the-art methods for image captioning that make use of the information extracted from knowledge graphs can substantially outperform those that rely solely on the information extracted from images., Comment: Accepted by WACV'19