In order to meet the demand for flexible acquisition of videoimages and real-timeprocessing of images, the speed and efficiency of imageprocessing are improved. This paper is an embedded system based on the ZYNQ-70...
详细信息
This research presents a real-time automotive sensing system for the data of urban garbage disposal. The proposed solution is implemented on an edge computing device mounted on garbage truck where a deep learning base...
详细信息
A real-time optical visualization system of the venous bed is presented, created using near infrared (NIR) vein finder technology with spectral division of the light flux (visible and infrared), using a combination of...
详细信息
Current Dynamic Texture Synthesis (DyTS) models can synthesize realistic videos. However, they require a slow iterative optimization process to synthesize a single fixed-size short video, and they do not offer any pos...
详细信息
ISBN:
(纸本)9798350301298
Current Dynamic Texture Synthesis (DyTS) models can synthesize realistic videos. However, they require a slow iterative optimization process to synthesize a single fixed-size short video, and they do not offer any post-training control over the synthesis process. We propose Dynamic Neural Cellular Automata (DyNCA), a framework for real-time and controllable dynamic texture synthesis. Our method is built upon the recently introduced NCA models and can synthesize infinitely long and arbitrary-sized realistic video textures in realtime. We quantitatively and qualitatively evaluate our model and show that our synthesized videos appear more realistic than the existing results. We improve the SOTA DyTS performance by 2 similar to 4 orders of magnitude. Moreover, our model offers several real-timevideo controls including motion speed, motion direction, and an editing brush tool. We exhibit our trained models in an online interactive demo that runs on local hardware and is accessible on personal computers and smartphones.
The moving target segmentation (MTS) aims to segment out moving targets in the video, however, the classical algorithm faces the huge challenge of real-timeprocessing in the current video era. Some scholars have succ...
详细信息
The moving target segmentation (MTS) aims to segment out moving targets in the video, however, the classical algorithm faces the huge challenge of real-timeprocessing in the current video era. Some scholars have successfully demonstrated the quantum advantages in some videoprocessing tasks, but not concerning moving target segmentation. In this paper, a quantum moving target segmentation algorithm for grayscale video is proposed, which can use quantum mechanism to simultaneously calculate the difference of all pixels in all adjacent frames and then quickly segment out the moving target. In addition, a feasible quantum comparator is designed to distinguish the grayscale values with the threshold. Then several quantum circuit units are designed in detail to construct the complete quantum circuits for segmenting the moving target. For a quantum video with 2m frames (every frame is a 2(n) x 2(n)$ image with q grayscale levels), the complexity of our algorithm can be reduced to O(n(2) + q). Compared with the classic counterpart, it is an exponential speedup, while its complexity is also superior to the existing quantum algorithms. Finally, the experiment is conducted on IBM Quantum Experience (IBM Q) to show the feasibility of our algorithm in the noisy intermediate-scale quantum era. In this paper, a quantum moving target segmentation algorithm for grayscale video is proposed, which can use quantum mechanisms to simultaneously process all pixels in the video and segment out the moving target. In addition, a quantum comparator with lower quantum cost and a complete circuit for quantum video segmentation is designed. The circuit complexity and experiment analysis demonstrate the superiority and feasibility of the ***
The article discusses the challenges of real-time data processing and analyzes various methods used to solve them, with a focus on imageprocessing. It points out the limitations of existing methods and argues for the...
详细信息
ISBN:
(纸本)9781510672895;9781510672888
The article discusses the challenges of real-time data processing and analyzes various methods used to solve them, with a focus on imageprocessing. It points out the limitations of existing methods and argues for the need to use more effective and modern technologies, proposing parallel- hierarchical networks as a promising solution. The article provides a detailed description of the structural-functional model of this type of network, which involves cyclically transforming the input data matrix using a "common part" criterion and an array evolution operator until a set of individual elements is formed. The proposed model is expected to improve real-timeimage recognition and can potentially be applied to other fields by using the "common part" criterion.
In the past decade, various haze removal techniques have been widely reported for object recognition. But hitherto little has been identified on the use of single image dehazing using transfer learning approach for ob...
详细信息
Due to the prohibitive cost as well as technical challenges in annotating ground-truth optical flow for large-scale realistic video datasets, the existing deep learning models for optical flow estimation mostly rely o...
详细信息
Due to the prohibitive cost as well as technical challenges in annotating ground-truth optical flow for large-scale realistic video datasets, the existing deep learning models for optical flow estimation mostly rely on synthetic data for training, which in turn may lead to significant performance degradation under test-data distribution shift in real-world environments. In this work, we propose the methodology to tackle this important problem. We design a self-supervised learning task for adjusting the optical flow estimation model at test time. We exploit the fact that most videos are stored in compressed formats, from which compact information on motion, in the form of motion vectors and residuals, can be made readily available. We formulate the self-supervised task as motion vector prediction, and link this task to optical flow estimation. To the best of our knowledge, our Test-time Adaption guided with Motion Vectors (TTA-MV), is the first work to perform such adaptation for optical flow. The experimental results demonstrate that TTA-MV can improve the generalization capability of various well-known deep learning methods for optical flow estimation, such as FlowNet, PWCNet, and RAFT.
Object detection from apron surveillance video is facing enormous storage pressure and computing overhead. Large cloud server cluster is generally used and high-speed network bandwidth is required, also equipped with ...
详细信息
Object detection from apron surveillance video is facing enormous storage pressure and computing overhead. Large cloud server cluster is generally used and high-speed network bandwidth is required, also equipped with powerful GPUs for computing support. The design of hardware-friendly and efficient object detection model is challenging. This paper presents a compression method for outdoor apron surveillance videos, which is further combined with a lightweight detection model to make the inference process independent of GPU. First, the gray level variance of dynamic objects is leveraged to binarize the monitoring images, then an improved MobileNet-SSD algorithm is proposed. Moreover, int8 quantization is performed and bit operations are designed to eliminate the floating-point operation and it can simultaneously accelerate and compress CNN models with only minor performance degradation. Experiment results on a large-scale dataset containing 22k monitoring images demonstrate that the compression ratio of quantized image can achieve up to 21 times, combined with quantized model, the detection on apron surveillance images can reach nearly 25FPS in a pure CPU environment, the mAP is 86.83%, and the model size is compressed to 600 kb. Significantly reduced computational complexity can be applied to embedded devices.
The Personal Data Protection Act (PDPA) was created to prevent the breach of personal information of users of computer systems without the data owner's consent. One type of data that frequently has problems with p...
详细信息
ISBN:
(纸本)9798350381771;9798350381764
The Personal Data Protection Act (PDPA) was created to prevent the breach of personal information of users of computer systems without the data owner's consent. One type of data that frequently has problems with privacy violations is images and videos. Because of difficult control, as a result, there are often extraneous people in the frame instead of just the intended subject. If the person caught in the frame does not want this information published, there will be a problem with that video. This causes the identity of the person to be concealed so that they can be identified. Although doing this is an acceptable method, censorship is labor-intensive and time-consuming. For these reasons, we proposed the Automated Face Selection and Censoring on image and video System using Multi-Task Cascaded Convolutional Neural Networks (MTCNN) Model to automatically detect the face and censor only unwanted persons. Furthermore, our proposed method also executes an automated system that can sense and ignore some frames that are not essential or redundant with other nearby frames to reduce complex processing and time-consumption.
暂无评论