Abstract

Monitoring water flow helps to identify leaks and wastage, leading to better management of water resources and conservation of this precious resource. To address this challenge, there is a need for an efficient and sustainable water management system. This paper presents an Internet of Things (IoT) based solution that involves retrofitting existing analog water meters using readily available off-the-shelf electronic components. Real-time data collection and analysis are performed through edge computation, which locally processes water meter images captured by the camera and extracts water meter readings. These readings are transmitted to the cloud for storage and further analysis. Various strategies have been implemented to optimize supply-current usage, preserving charge-discharge cycles of solar-powered batteries even in adverse environmental conditions. To streamline the firmware update process for multiple connected devices, a broadcasting technique is employed, offering the benefits of reduced manual labor and time savings. To assess the reliability and performance of developed solution, field deployment is conducted over several months, enabling the characterization of water usage patterns across different locations. Integrating energy harvesting capabilities into system reduces maintenance costs and promotes eco-friendly energy practices. Overall, this solution offers an effective and comprehensive approach towards achieving efficient and sustainable water management.

Abstract

Particulate matter (PM) is considered the primary contributor to air pollution and has severe implications for general health. PM concentration has high spatial variability and thus needs to be monitored locally. Traditional PM monitoring setups are bulky, expensive, and cannot be scaled for dense deployments. This paper argues for a densely deployed network of IoT-enabled PM monitoring devices using low-cost sensors, specifically focusing on PM10 and PM2.5, the most health-impacting particulates. In this work, 49 devices were deployed in a region of the Indian metropolitan city of Hyderabad, of which 43 devices were developed as part of this work, and six devices were taken off the shelf. The low-cost sensors were calibrated for seasonal variations using a precise reference sensor and were particularly adjusted to accurately measure PM10 and PM2.5 levels. A thorough analysis of data collected for 7 months has been presented to establish the need for dense deployment of PM monitoring devices. Different analyses such as mean, variance, spatial interpolation, and correlation have been employed to generate interesting insights about temporal and seasonal variations of PM10 and PM2.5. In addition, event-driven spatio-temporal analysis is done for PM2.5 and PM10 values to understand the impact of the bursting of firecrackers on the evening of the Diwali festival. A web-based dashboard is designed for real-time data visualization.

Abstract

Autonomous Underwater Vehicles (AUVs) play a vital role in explor- ing and mapping underwater environments. However, the presence of nadir gaps, or blind zones, in commercial AUVs can lead to unex- plored areas during mission execution, limiting their effectiveness. Our work addresses the challenges of path planning in the presence of nadir gaps and presents scalable coverage strategies for AUVs minimizing either the mission completion time or the total number of turns performed while ensuring complete exploration, eliminating the risk of leaving critical areas unexplored. We provide provably complete strategies and perform extensive simulations on diverse input configurations based on real-world instances to demonstrate the efficacy of our strategies.

Abstract

Edge server-assisted computation offloading enables vehicles to leverage server compute resources to deliver connected services, overcoming the limitations of onboard resources. Understanding the compute workloads of edge servers is crucial for effective resource management and scheduling, yet this task is challenging due to the complex interplay of factors such as vehicle mobility and computation offloading patterns. To address this, we propose an empirical analysis framework that systematically characterizes the compute workloads of edge servers. We begin by formalizing the relationships among three key aspects: local load (generated by vehicles), composite load (imposed on edge servers), and traffic flow (vehicle mobility patterns). Our framework then uses models of the local load and traffic flow as inputs to generate the composite loads on edge servers. Experiments were conducted by injecting between 600 and 5,000 vehicles per hour in two distinct geographical areas, New York City and Tampa. We provide a quantitative analysis demonstrating how the composite loads on edge servers vary with changes in traffic flows, geographical areas, and offloading patterns.

Abstract

An agent assisting humans in daily living activities can collaborate more effectively by anticipating upcoming tasks. Data-driven methods represent the state of the art in task anticipation, planning, and related problems, but these methods are resource-hungry and opaque. Our prior work introduced a proof of concept framework that used an LLM to anticipate 3 high-level tasks that served as goals for a classical planning system that computed a sequence of low-level actions for the agent to achieve these goals. This paper describes DaTAPlan, our framework that significantly extends our prior work toward human-robot collaboration. Specifically, DaTAPlan’s planner computes actions for an agent and a human to collaboratively and jointly achieve the tasks anticipated by the LLM, and the agent automatically adapts to unexpected changes in human action outcomes and preferences. We evaluate DaTAPlan’s capabilities in a realistic simulation environment, demonstrating accurate task anticipation, effective human-robot collaboration, and the ability to adapt to unexpected changes.

Abstract

Certain self-supervised approaches to train image encoders, like CLIP, align images with their text captions. However, these approaches do not have an a priori incentive to learn to associate text inside the image with the semantics of the text. Humans process text visually; our work studies the semantics of text rendered in images. We show that the semantic information captured by image representations can decisively classify the sentiment of sentences and is robust against visual attributes like font and not based on simple character frequency associations.

Abstract

We present a method to compress the final linear layer of language models, reducing memory usage by up to 3.4 x without significant performance loss. By grouping tokens based on Byte Pair Encoding (BPE) merges, we prevent materialisation of the memory-intensive logits tensor. Evaluations on the TinyStories dataset show that our method performs on par with GPT-Neo and GPT2 while significantly improving throughput by up to 3x, making it suitable for low-compute environments.

Abstract

Forensic sketch-to-mugshot matching is a challenging task in face recognition, primarily hindered by the scarcity of annotated forensic sketches and the modality gap between sketches and photographs. To address this, we propose CLIP4Sketch, a novel approach that leverages diffusion models to generate a large and diverse set of sketch images, which helps in enhancing the performance of face recognition systems in sketch-to-mugshot matching. Our method utilizes Denoising Diffusion Probabilistic Models (DDPMs) to generate sketches with explicit control over identity and style. We combine CLIP and Adaface embeddings of a reference mugshot, along with textual descriptions of style, as the conditions to the diffusion model. We demonstrate the efficacy of our approach by generating a comprehensive dataset of sketches corresponding to mugshots and training a face recognition model on our synthetic data. Our results show significant improvements in sketch-to-mugshot matching accuracy over training on an existing, limited amount of real face sketch data, validating the potential of diffusion models in enhancing the performance of face recognition systems across modalities. We also compare our dataset with datasets generated using GAN-based methods to show its superiority.

Abstract

Portrait sketching involves capturing identity specific attributes of a real face with abstract lines and shades. Unlike photo-realistic images, a good portrait sketch generation method needs selective attention to detail, making the problem challenging. This paper introduces textbf{Portrait Sketching StyleGAN (PS-StyleGAN)}, a style transfer approach tailored for portrait sketch synthesis. We leverage the semantic $W+$ latent space of StyleGAN to generate portrait sketches, allowing us to make meaningful edits, like pose and expression alterations, without compromising identity. To achieve this, we propose the use of Attentive Affine transform blocks in our architecture, and a training strategy that allows us to change StyleGAN's output without finetuning it. These blocks learn to modify style latent code by paying attention to both content and style latent features, allowing us to adapt the outputs of StyleGAN in an inversion-consistent manner. Our approach uses only a few paired examples ($sim 100$) to model a style and has a short training time. We demonstrate PS-StyleGAN's superiority over the current state-of-the-art methods on various datasets, qualitatively and quantitatively.

Abstract

Edge computing allows for reduced latency and operational costs compared to centralized cloud systems. In this context, serverless functions are emerging as a lightweight and effective paradigm for managing computational tasks on edge infrastructures. However, the placement of such functions in constrained edge nodes remains an open challenge. On one hand, it is key to minimize network delays and optimize resource consumption; on the other hand, decisions must be made in a timely manner due to the highly dynamic nature of edge environments. In this paper, we propose POSEIDON, a solution based on Deep Reinforcement Learning for the efficient placement of functions at the edge. POSEIDON leverages Proximal Policy Optimization (PPO) to place functions across a distributed network of nodes under highly dynamic workloads. A comprehensive empirical evaluation demonstrates that POSEIDON significantly reduces execution time, network delay, and resource consumption compared to state-of-the-art methods.