the proceedings contain 15 papers. the topics discussed include: self-supervised 3D face reconstruction based on dense key points;slice-based ray casting volume shadow of volumetric datasets;a GPU computation-based ra...
ISBN:
(纸本)9781510679771
the proceedings contain 15 papers. the topics discussed include: self-supervised 3D face reconstruction based on dense key points;slice-based ray casting volume shadow of volumetric datasets;a GPU computation-based ray tracing engine with user-friendly and scalable rendering features and structures;a lightweight stereo depth estimation network based on mobile devices;a lightweight stereo depth estimation network based on mobile devices;grand challenge of imageprocessing in automatic detection of vehicles running in red lights;anomaly detection algorithm for asymmetric autoencoder based on knowledge distillation;a stereo vision-based real-time 3D hand pose estimation system combining nonlinear optimization;and a lightweight real-time 3D hand gesture tracking solution for mobile devices.
Deep learning models for computervision in remote sensing such as Convolutional Neural Network (CNN) has benefitted acceleration from the usage of multiple CPUs and GPUs. there are several ways to make the training s...
详细信息
ISBN:
(纸本)9781510666184;9781510666191
Deep learning models for computervision in remote sensing such as Convolutional Neural Network (CNN) has benefitted acceleration from the usage of multiple CPUs and GPUs. there are several ways to make the training stage more effective in terms of utilizing multiple cores at the same time by processing different image mini-batches with a duplicated model called Distributed Data Parallelization (DDP) and computing the parameters in a lower precision floating-point number called Automatic Mixed Precision (AMP). We would like to investigate the impact of DDP and AMP training modes on the overall utilization and memory consumption of CPU and GPU, as well as the accuracy of a CNN model. the study is performed on the EuroSAT dataset, a Sentinel-2-based benchmark satellite image dataset for image classification of land covers. We compare training using 1 CPU, using DDP, and using both DDP and AMP over 100 epochs using ResNet-18 architecture. the hardware that we used are Intel Xeon Silver 4116 CPU with 24 cores and an NVIDIA v100 GPU. We find that although parallelization of CPUs or DDP takes less time to train on the images, it can take 50 MB more memory than using only a single CPU. the combination of DDP and AMP can release memory up to 160 MB and reduce computation time by 20 seconds. the test accuracy is slightly higher for both DDP and DDP-AMP at 90.61% and 90.77% respectively than without DDP and AMP at 89.84%. Hence, training using Distributed Data Parallelization (DDP) and Automatic Mixed Precision (AMP) has more benefits in terms of lower GPU memory consumption, faster training execution time, faster convergence towards solutions, and finally, higher accuracy.
Video Action Recognition (VAR) is a challenging task due to its inherent complexities. though different approaches have been explored in the literature, designing a unified framework to recognize a large number of hum...
详细信息
Satellite image super resolution is an important task that generates high resolution satellite images from low resolution inputs. Multi-frame super resolution utilizes multiple low-resolution images to generate a sing...
详细信息
Domain shifts are a common problem in computervision. As a result, a classifier trained on a source domain cannot perform well on a target domain. Due to this, a source classifier trained to differentiate based on a ...
详细信息
A mathematical model of controlled motion (called the Dubins car in the literature on optimal control) is considered. this model is widely used to describe various motions: an airplane in a horizontal plane, a car, et...
详细信息
In the realm of computervision, image denoising remains a formidable challenge with profound implications for fields like medical imaging, remote sensing, and photography. Despite notable advancements in deep learnin...
详细信息
Identifying and locating objects in images and videos, including elements like traffic signs, vehicles, buildings, and people, constitutes a fundamental and demanding task in computervision, known as object detection...
详细信息
ISBN:
(纸本)9783031821523;9783031821530
Identifying and locating objects in images and videos, including elements like traffic signs, vehicles, buildings, and people, constitutes a fundamental and demanding task in computervision, known as object detection. Due to the higher computing complexity of this technique and the large amount of data carried by the video signal, it is nearly impossible for ordinary general-purpose processors GPPs or CPUs to run these techniques in real-time, especially for embedded systems applications. therefore, special hardware that can acquire, control, or execute in parallel is required. these specialized hardware systems include Digital Signal Processors DSPs, Field Programmable Gate Arrays FPGAs, Visual processing Units VPUs, Tensor processing Units TPUs, Neural processing Units NPUs or graphicsprocessing Units GPUs. this work presents the benefits of accelerating traditional object detection methods on a high-end embedded system, the Jetson Nano Developer Kit. this single computer board is equipped withthe Tegra K1 System on Chip SoC, which is composed of a quad-core ARM A15 and 192 cores of Kepler-embedded GPU. Computing acceleration was ensured via the use of the CUDA OpenCV library for boththe Histogram of Oriented Gradients HOG and the Haar Cascade Classifier. For VGA resolution, results reveal that the GPU implementation on this embedded system is 1.4x faster than the CPU for the HOG method and 2x for the Haar Cascade Classifier method.
We focus on domain and class generalization problems in analyzing optical remote sensing images, using the large-scale pre-trained vision-language model (VLM), CLIP. While contrastively trained VLMs show impressive ze...
详细信息
this paper introduces a novel technique of computational art with mandala-an iconic heritage of indian folk art. Its novelty lies in several fundamental steps. the first one is fixing the asymmetries and the imperfect...
详细信息
暂无评论