Object detection models often struggle with class imbalance, where rare categories appear significantly less frequently than common ones. Existing sampling-based rebalancing strategies, such as Repeat Factor Sampling (RFS) and Instance-Aware Repeat Factor Sampling (IRFS), mitigate this issue by adjusting sample frequencies based on image and instance counts. However, these methods are based on linear adjustments, which limit their effectiveness in long-tailed distributions. This work introduces Exponentially Weighted Instance-Aware Repeat Factor Sampling (E-IRFS), an extension of IRFS that applies exponential scaling to better differentiate between rare and frequent classes. E-IRFS adjusts sampling probabilities using an exponential function applied to the geometric mean of image and instance frequencies, ensuring a more adaptive rebalancing strategy. We evaluate E-IRFS on a dataset derived from the Fireman-UAV-RGBT Dataset and four additional public datasets, using YOLOv11 object detection models to identify fire, smoke, people and lakes in emergency scenarios. The results show that E-IRFS improves detection performance by 22% over the baseline and outperforms RFS and IRFS, particularly for rare categories. The analysis also highlights that E-IRFS has a stronger effect on lightweight models with limited capacity, as these models rely more on data sampling strategies to address class imbalance. The findings demonstrate that E-IRFS improves rare object detection in resource-constrained environments, making it a suitable solution for real-time applications such as UAV-based emergency monitoring. The code is available at: https://github.com/futurians/E-IRFS.

Image live co-editing (LCE), which allows users to edit a shared image concurrently and remotely, is rising in popularity. However, fluctuating resources (i.e., bandwidth and computation), as well as varying degrees of edit complexity, make it challenging to achieve low-latency image live co-editing, which drastically degrades the user experience. To address this issue, we propose Alice, a cross-platform compression adaptation framework that incorporates three core designs. First, Alice leverages both data-based (i.e., sending compressed pixels) and operation-based (i.e., sending image editing operation APIs and corresponding parameters) approaches for image edit transmission. Second, Alice combines diverse modern lossless compression techniques and their various configurations to enhance the adaptability of data-based transmission. Third, Alice features a lookup table (LUT)-based decision framework to determine the best transmission strategy for image edits in real time. We implement Alice and integrate it into our image LCE testbed. Our extensive evaluation shows that, compared to the baselines using a fixed transmission strategy, Alice achieves up to 95% latency reduction with negligible overhead.

We present NIER, a video conferencing system that can adaptively maintain a low bitrate (e.g., 10–100 Kbps) with reasonable visual quality while being robust to packet losses. We use key-point-based deep image animation (DIA) as a key building block and address a series of networking and system challenges to make NIER practical. Our evaluations show that NIER significantly outperforms the baseline solutions.

Video Conferencing Applications (VCAs) are indispensable for real-time communication in remote work and education by enabling simultaneous transmission of audio, video, and screen-sharing content. Despite their ubiquity, research on how these platforms allocate network bandwidth, especially under constrained conditions, and how these resource allocation strategies affect the users' Quality of Experience (QoE) is lacking. This paper addresses this gap by analyzing bandwidth allocation strategies in Zoom, Webex, and Google Meet, with a focus on QoE implications. To assess QoE, we propose a general QoE prediction model based on data collected from a study involving 800 participants. This study is a pioneering effort in evaluating multimedia transmissions across diverse scenarios and network conditions, advancing beyond prior research focused on single media types. The results demonstrate the model's effectiveness and generality in predicting QoE across various VCA scenarios.

Volumetric videos offer a unique interactive experience and have the potential to enhance social virtual reality and telepresence. Streaming volumetric videos to multiple users remains a challenge due to its tremendous requirements of network and computation resources. In this paper, we develop MuV2, an edge-assisted multi-user mobile volumetric video streaming system to support important use cases such as tens of students simultaneously consuming volumetric content in a classroom. MuV2 achieves high scalability and good streaming quality through three orthogonal designs: hybridizing direct streaming of 3D volumetric content with remote rendering, dynamically sharing edge-transcoded views across users, and multiplexing encoding tasks of multiple transcoding sessions into a limited number of hardware encoders on the edge. MuV2 then integrates the three designs into a holistic optimization framework. We fully implement MuV2 and experimentally demonstrate that MuV2 can deliver high-quality volumetric videos to over 30 concurrent untethered mobile devices with a single WiFi access point and a commodity edge server.

Collaborative Vehicular Perception (CVP) enables connected and autonomous vehicles (CAVs) to cooperatively extend their views through wirelessly sharing their sensor data. Existing CVP systems employ either a vehicle-to-vehicle (V2V) or vehicle-to-infrastructure (V2I) view exchange paradigm. In this paper, we advocate a hybrid CVP design: our developed system, Harbor, employs V2I as its fundamental underlying framework, and opportunistically employs V2V to boost the performance. In Harbor, vehicles (helpers) may serve as relays to assist other vehicles (helpees) in reaching an edge node, which performs sensor data merging to produce the extended view. We judiciously partition the workload between the edge and vehicles, develop a robust helper-helpee assignment model, and solve it efficiently at runtime. We conduct both real-world tests and large-scale emulation experiments using two prevailing CAV applications: drivable space detection and object detection. Our real-world evaluation conducted at one of the world's first purpose-built autonomous driving testbeds demonstrates that Harbor outperforms state-of-the-art V2V- or V2I-only CVP schemes by up to 36% in detection accuracy, resulting in significantly fewer collisions under dangerous driving scenarios.

By aggregating multiple channels, Carrier Aggregation (CA) is an important technology for boosting cellular network bandwidth. Given diverse radio bands made available in 5G networks, CA plays a particularly critical role in achieving the goal of multi-Gbps throughput performance. In this paper, we carry out a timely comprehensive measurement study of CA deployment in commercial 5G networks (as well as 4G networks). We identify the key factors that influence whether CA is deployed and when, as well as which band combinations are used. Thus, we reveal the challenges posed by CA in 5G performance analysis and prediction as well as their implications in application quality-of-experience (QoE). We argue for and develop a novel CA-aware deep learning framework, dubbed Prism5G, which explicitly accounts for the complexity introduced by CA to more effectively predict 5G network throughput performance. Through extensive evaluations, we demonstrate the superiority of Prism5G over existing throughput prediction algorithms. Prism5G improves 5G throughput prediction accuracy by over 14% on average and a maximum of 22%. Using two use cases as examples, we further illustrate how Prism5G can aid applications in optimizing QoE performance.

With 5G's support for diverse radio bands and different deployment modes, e.g., standalone (SA) vs. non-standalone (NSA), mobility management - especially the handover process - becomes far more complex. Measurement studies have shown that frequent handovers cause wild fluctuations in 5G throughput, and worst, service outages. Through a cross-country (6,200 km+) driving trip, we conduct in-depth measurements to study the current 5G mobility management practices adopted by three major U.S. carriers. Using this rich dataset, we carry out a systematic analysis to uncover the handover mechanisms employed by 5G carriers, and compare them along several dimensions such as (4G vs. 5G) radio technologies, radio (low-, mid- & high-)bands, and deployment (SA vs. NSA) modes. We further quantify the impact of mobility on application performance, power consumption, and signaling overheads. We identify key challenges facing today's NSA 5G deployments which result in unnecessary handovers and reduced coverage. Finally, we design a holistic handover prediction system Prognos and demonstrate its ability to improve QoE for two 5G applications 16K panoramic VoD and realtime volumetric video streaming. We have released the artifacts of our study at https://github.com/SIGCOMM22-5GMobility/artifact.

Connected and Autonomous Vehicles (CAVs) heavily rely on 3D sensors such as LiDARs, radars, and stereo cameras. However, 3D sensors from a single vehicle suffer from two fundamental limitations: vulnerability to occlusion and loss of details on far-away objects. To overcome both limitations, in this paper, we design, implement, and evaluate EMP, a novel edge-assisted multi-vehicle perception system for CAVs. In EMP, multiple nearby CAVs share their raw sensor data with an edge server which then merges CAVs' individual views to form a more complete view with a higher resolution. The merged view can drastically enhance the perception quality of the participating CAVs. Our core methodological contribution is to make the sensor data sharing scalable, adaptive, and resource-efficient over oftentimes highly fluctuating wireless links through a series of novel algorithms, which are then integrated into a full-fledged cooperative sensing pipeline. Extensive evaluations demonstrate that EMP can achieve real-time processing at 24 FPS and end-to-end latency of 93 ms on average. EMP reduces the end-to-end latency by 49% to 65% compared to the traditional vehicle-to-vehicle (V2V) sharing approach without edge support. Our case studies show that cooperative sensing powered by EMP can detect hazards such as blind spots faster by 0.5 to 1.1 seconds, compared to a single vehicle's perception.

Volumetric videos allow viewers to exercise 6-DoF (degrees of freedom) movement when watching them. Due to their true 3D nature, streaming volumetric videos is highly bandwidth demanding. In this work, we present to our knowledge a first volumetric video streaming system that leverages deep super resolution (SR) to boost the video quality on commodity mobile devices. We propose a series of judicious optimizations to make SR efficient on mobile devices.

Deep neural networks (DNNs) are vulnerable to adversarial examples where inputs with imperceptible perturbations mislead DNNs to incorrect results. Recently, adversarial patch, with noise confined to a small and localized patch, emerged for its easy accessibility in real-world. However, existing attack strategies are still far from generating visually natural patches with strong attacking ability, since they often ignore the perceptual sensitivity of the attacked network to the adversarial patch, including both the correlations with the image context and the visual attention. To address this problem, this paper proposes a perceptual-sensitive generative adversarial network (PS-GAN) that can simultaneously enhance the visual fidelity and the attacking ability for the adversarial patch. To improve the visual fidelity, we treat the patch generation as a patch-to-patch translation via an adversarial process, feeding any types of seed patch and outputting the similar adversarial patch with high perceptual correlation with the attacked image. To further enhance the attacking ability, an attention mechanism coupled with adversarial generation is introduced to predict the critical attacking areas for placing the patches, which can help producing more realistic and aggressive patches. Extensive experiments under semi-whitebox and black-box settings on two large-scale datasets GTSRB and ImageNet demonstrate that the proposed PS-GAN outperforms state-of-the-art adversarial patch attack methods.