Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 71 for “"Real-Time Inference"”.
-
Energy-Efficient Recurrent Neural Network Accelerators for Real-Time Inference
… Neural Networks (RNN) are helpful in processing time-sequential data and are widely used in applications such as speech recognition and machine translation. RNNs are difficult to compute because of their massive arithmetic operations and large memory footprint. RNN inference workloads used to be …
-
Real-Time Inference of Topological Structure and Vulnerabilities for Adaptive Jamming Against Covert Ad Hoc Networks
… synchronicity of transmission timings in both time and frequency domains, complemented with a minimal thresholding method for classification of link estimations. Furthermore, this work proposes a computationally efficient method for identification of the most vulnerable region of the network …
-
Bayesian edge analytics of machine process and health status in an IoT framework
… in size to facilitate health monitoring and near real-time inference. The computational latency of this proposed methodology compares favorably to cloud-based solutions, where network transmission latency introduces significant variance in obtaining statistical features and model inference. A case …
-
Fast and Data-Efficient Image Segmentation
… require human annotation of data and exhibits real-time inference. First, we introduce the use of residual neural networks for surgical instrument segmentation for robotic surgery. We show state of the art results on multiple instrument segmentation datasets. Second, we introduce a neural …
-
Real Time People Segmentation with 10% CPU Usage for Video Conference
… frame skipping techniques, leading to a fast inference speed on CPU. Our extensive experiments show that the proposed CEPeopleSeg can achieve a high prediction mIoU of 87.9% on Supervisely People Dataset while reaching a real-time inference speed of 32.40 fps on CPU with very low usage of 10%.
-
Multi-Task Learning For Deployable, Boundary-Refined Semantic Segmentation
… Low-level feature preservation and robust, real-time network deployment are regarded as two of the biggest challenges in this task. The former challenge loses low-level features during downsampling operations, which are integral to learning object semantics, posing a seemingly paradoxical …
-
TOWARDS EFFICIENT AND GENERALIZABLE 3D HUMAN POSE ESTIMATION AND ITS APPLICATIONS
… body mesh estimation, which achieves real-time inference speed. Furthermore, we present an inference stage optimization pipeline, which helps improve model generalization ability under out-of-distribution scenes significantly. Lastly, we adopt 3D human poses as priors and develop a 3D …
-
Toward Robust Semantic Segmentation in Levee Infrastructure Monitoring: Enhancing Accuracy with High-Fidelity Synthetic Data and Ensemble Learning
… adds structural constraints to enhance realism and consistency. A semi-automatic convex hull annotation technique enhances labeling efficiency, and ensemble learning strategies further improve segmentation accuracy. The system integrates real-time inference capabilities within a …
-
Design of an 8-bit Configurable CNN Hardware Accelerator for Audio and Image Classification
… CNNs in low-power consumption hardware for real-time inference. This research aims to design a configurable hardware accelerator for 8-bit fixed point audio and image CNN models. An audio network is developed to classify environmental sounds from children playing in the streets, car horns, …
-
Rethinking Serverless for Machine Learning Inference
… intelligence and machine learning, AI/ML inference tasks have become exceedingly popular. However, executing these workloads on dedicated hardware may not be feasible for many users due to high maintenance costs, varying load patterns, and time to production. Furthermore, ML inference …
-
Neural Manifolds, Just in Time: Real-time Dimensionality Reduction for Neural Populations
… reduction methods to address challenges in the real-time inference and prediction of low-dimensional neural dynamics, Chapter 3 and Chapter 4 motivate and present novel models for inferring temporally sparse dynamics to explore the idea of dimensionality in a framework of low-dimensional neural …
-
Retinomorphic Event-Driven Representations for Video Tasks
… flow. EDR's fast processing speed enables real-time inference/learning in time sensitive video applications such as reinforcement learning. In this vein, we use EDR to demonstrate performance improvements over state-of-the-art reinforcement learning algorithms for Atari games, something …
-
MODEL ADAPTATION FOR EDGE AI
… DNN models can lead to wasteful and inefficient inference. We propose CRISP, a class-aware pruning framework that employs a hybrid structured sparsity pattern. Thirdly, while low-precision integer formats are common for reduced precision arithmetic, these formats are inadequate for emerging …
-
Efficient Processing of Convolutional Neural Networks on the Edge: A Hybrid Approach Using Hardware Acceleration and Dual-Teacher Compression
… that maintain high accuracy and perform real-time inference. Driven by open-source hardware design frameworks such as FINN and HLS4ML, this research focuses on hardware acceleration, model compression, and efficient implementation of CNN algorithms on AMD SoC-FPGAs using High-Level …
-
MODEL OPTIMIZATION FOR VEHICLE RECOGNITION ON EDGE DEVICES
… and power constraints while maintaining good inference performance. This research explores ML for vehicle recognition via transfer learning of various state-of-the-art convolutional neural network models. Also, we study the effects of applying optimization techniques, pruning, and …
-
Data-Driven Methods for Reduced Order Models, System Identification, and Feedback Control
… dynamical systems from limited, noisy, and real-world data. We introduce a reduced-order modeling framework to capture the initial ejection dynamics of turbulent plumes directly from video. By combining Otsu's thresholding with a concentric-circle search to extract plume geometries, we …
-
A Power System-Based IoT Network for Remote Sensing Applications
… aided with widespread data collection and real-time inference. However, as IoT networks scale and require more energy for near-data analytics, the IoT endpoints require more power and complexity, limiting their deployment. Additionally, deploying endpoints in remote locations creates …
-
Autonomous Sensor Tasking for Space Situational Awareness using Deep Reinforcement Learning
… physically-based machine learning algorithm for real-time inference of Space Objects energy parameters and states to predict the space objects orbits for the purpose of sensor tasking system to meet the high-level goals of Space Situational Awareness. By using a predictive algorithm, the system …
-
Translating Human Emotion into Music: A Multimodal Machine-Learning Framework Linking Movement, Facial Expression, and Real-Time Generative Composition
… into musically coherent behaviour using time-series machine learning and real-time generative systems. The project develops and evaluates a multimodal framework that links participant-specific emotion inference to an interpretable, artist-configurable generative music system.Across two …
-
Supervised and Unsupervised Deep Learning Methods for Underwater Image Enhancement
… and color casting on underwater imagery in real time. We use two parallel architectures trained in a generative adversarial network scheme (GAN) with channel and spatial attention blocks to retrieve color, and discrete wavelength transform to preserve high-frequency components. Our …
Page 1 of 4