Recent advances in large video-language models have revolutionized video understanding tasks.However, their efficiency is greatly constrained by processing high volumes of visual tokens.Existing token compression strategies apply a fixed compression ratio, ignoring varying semantic density across video clips.Consequently, this leads to inadequate representation of information-rich clips due to insufficient tokens and unnecessary computation on static or content-poor ones.To address this, we propose LangDC, a Language-aware Dynamic Token Compressor.LangDC leverages a lightweight language model to describe video clips, converting them into soft caption tokens as visual representations.Trained with our proposed semantic density-aware supervision, LangDC aims to 1) cover key visual cues necessary for downstream task reasoning and 2) dynamically adjust compression ratios based on scene richness, reflected by description length.Our design mimics how humans dynamically express what they see: complex scenes (seeing more) elicit more detailed language to convey nuances (saying more), whereas simpler scenes are described with fewer words.Experimental results show that our method reduces FLOPs by 49% compared to VideoGPT+ while maintaining competitive performance.Furthermore, qualitative results demonstrate our approach adaptively adjusts the token compression ratio based on video segment richness.Codes are available at https://github.com/NIneeeeeem/LangDC.

Abstract This paper focuses on the intelligent detection of prohibited items in X‐ray images during the security checking process. An intelligent semantic segmentation model of prohibited items in X‐ray images is proposed based on the attention‐based object localization method. Based on the pre‐trained CNN classification framework, the attention mechanism can map the high‐layer semantic information of objects into the input space, while generating energy saliency maps to locate the prohibited items. In order to make the obtained attention maps discriminative, the lateral and contrastive inhibition strategies are introduced and combined together which can highlight the responses of activated neurons. Under the guidance of attention responses, two traditional image segmentation algorithms are employed to achieve the semantic segmentation results for the prohibited items detection in X‐ray images. The proposed semantic segmentation model relies on weakly supervised learning mechanism, and only depends on the category labels of prohibited items, which greatly avoids the work cost of data semantic annotation. The experimental results based on the public SIXray baseline and the self‐built X‐ray image database demonstrate the proposed method can achieve about 65% IoU localization precise averagely. In addition, comparison experiments were carried out with the state‐of‐the‐arts and ablation experiments to verify the effectiveness of the proposed model.

Time sensitive network applications, for example in Intra-Vehicular Networks, aim to give predictable end-to-end latency guarantees. As a consequence, processing resources of involved host systems remain partially unused, because they are reserved for rare worst cases. This circumstance provides the opportunity to reduce dimensioning overheads by managing the load on the nodes flexibly within the network. In our proposed approach, a SmartNIC involving an FPGA-based load balancer achieves dynamic routing of flows whilst preserving end-to-end latency guarantees. A flow-oriented online network measurement component continuously supervises network traffic with regards to compliance to flow specifications and constraints such as bounded one-way delay, absence of packet loss, and jitter. We use the supervisor to enhance forwarding decisions on the data plane. Initial evaluation yields a saving potential of around 30 %. We showcase quick dynamic reconfiguration of the FPGA when triggered by real-time measurement of the one-way delay using realistic automotive network traffic.

Driver assistance, self-driving, and multimedia systems have two common implications: increasing demand on network bandwidth and the need for more powerful computation nodes. As a result, intra-vehicular networks (IVNs) change their layout. They are built around central nodes connected to the rest of the vehicle via Ethernet. The usage of Ethernet presents a challenge, as it lacks support for deterministic behavior by design. The solution is found within the IEEE Time-Sensitive Networking (TSN) standards, introducing real-time, low-latency, and deterministic communication into the Ethernet ecosystem. These new networked systems need to be thoroughly evaluated with IVN requirements in mind. To assess numerous configurations of IVN setups, in this work, we introduce a novel Environment for Generic In-vehicular Networking Experiments — EnGINE. It allows, among many others, repeatable, reproducible, and replicable TSN experiments with high precision and flexibility, which is not possible to run using proprietary solutions. EnGINE is based exclusively on commercial off-the-shelf components and is orchestrated by a flexible Ansible framework. This approach allows us to configure various topologies emulating realistic IVNs behavior, which is challenging using simulations. Based on available related work, we further address the challenges found in the IVNs. We derive additional requirements for experiments in the TSN domain and present our approach to fulfill them in an experimental setting. We believe that EnGINE provides the ideal environment for TSN network experiments.

Ethernet is increasingly used in areas where time-critical and safety-relevant data are transported over the network along with best-effort flows, for example in intra vehicle networks or industrial networks. The resulting complex network architectures, time-sensitive networking configurations and system interactions are hard to foresee during the design phase. Therefore, it is hard to rule out any violations of flow specifications or timing and reliability requirements, especially in the presence of unpredictable failures.

As a class of block codes, LDPC codes with any desired code rate and code length are easily constructed. We examine the performance of three high-rate (0.769, 0.889, 0.935) LDPC codes in image transmission. Simulation results show that LDPC codes are good coding schemes over fading channels in image communication with lower system complexity. The distorted images can be recovered by using the three high-rate LDPC codes at SNR of 10 dB, 11.5 dB and 14 dB respectively. These high-rate codes can achieve relatively higher bandwidth efficiency than the low-rate codes can. Moreover, we discovered that the code rate has more influence on the performance than the code length does.

Two adaptive coded modulation schemes employing LDPC (low-density parity-check) code are proposed for Rayleigh fading channels. Scheme 1: we fix the code rate (rate=1/2), and only change the modulation methods according to the channel state, that is, during poor channel conditions, QPSK modulation is employed, as channel conditions improve, more efficient modulation scheme such as 8PSK is used. Scheme 2: the modulation scheme is fixed (BPSK modulation), and the code rate is changed according to the channel state. Through simulation, we get the performance of these two adaptive LDPC schemes under different demands of BER, and the comparison between adaptive LDPC and nonadaptive LDPC is also given.

Low-density parity check coded bit-interleaved coded modulation (BICM) schemes are analyzed in this paper. A simplified decoding method without iteration between demodulator and decoder is provided. And the performance over additive white Gaussian noise (AWGN) and Rayleigh fading channels are analyzed. Through the simulation results we can conclude that the schemes with low complexity decoding method have good performance both over AWGN and Rayleigh fading channels.