Deep Learning Based Point Cloud Processing And Compression
Download Deep Learning Based Point Cloud Processing And Compression PDF/ePub or read online books in Mobi eBooks. Click Download or Read Online button to get Deep Learning Based Point Cloud Processing And Compression book now. This website allows unlimited access to, at the time of writing, more than 1.5 million titles, including hundreds of thousands of titles in various foreign languages.
Deep Learning Based Point Cloud Processing and Compression
A point cloud is a 3D data representation that is becoming increasingly popular. Recent significant advances in 3D sensors and capturing techniques have led to a surge in the usage of 3D point clouds in virtual reality/augmented reality (VR/AR) content creation, as well as 3D sensing for robotics, smart cities, telepresence, and automated driving applications. With an increase in point cloud applications and improved capturing technologies, we now have high-resolution point clouds with millions of points per frame. However, due to the large size of a point cloud, efficient techniques for the transmission, compression, and processing of point cloud content are still widely sought. This thesis addresses multiple issues in the transmission, compression, and processing pipeline for point cloud data. We employ a deep learning solution to process 3D dense as well as sparse point cloud data for both static as well as dynamic contents. Employing deep learning on point cloud data which is inherently sparse is a challenging task. We propose multiple deep learning-based frameworks that address each of the following problems: Point Cloud Compression Artifact Removal. V-PCC is the current state-of-the-art for dynamic point cloud compression. However, at lower bitrates, there are unpleasant artifacts introduced by V-PCC. We propose a deep learning solution for V-PCC artifact removal by leveraging the direction of projection property in V-PCC to remove quantization noise. Point Cloud Geometry Prediction. The current point cloud lossy compression and processing techniques suffer from quantization loss which results in a coarser sub-sampled representation of the point cloud. We solve the problem of points lost during voxelization by performing geometry prediction across spatial scales using deep learning architecture. Point Cloud Geometry Upsampling. Loss of details and irregularities in point cloud geometry can occur during the capturing, processing, and compression pipeline. We present a novel geometry upsampling technique, PU-Dense, which can process a diverse set of point clouds including synthetic mesh-based point clouds, real-world high-resolution point clouds, real-world indoor LiDAR scanned objects, as well as outdoor dynamically acquired LiDAR-based point clouds. Dynamic Point Cloud Interpolation. Dense photorealistic point clouds can depict real-world dynamic objects in high resolution and with a high frame rate. Frame interpolation of such dynamic point clouds would enable the distribution, processing, and compression of such content. We also propose the first point cloud interpolation framework for photorealistic dynamic point clouds. Inter-frame Compression for Dynamic Point Clouds. Efficient point cloud compression is essential for applications like virtual and mixed reality, autonomous driving, and cultural heritage. We propose a deep learning-based inter-frame encoding scheme for dynamic point cloud geometry compression. In each case, our method achieves state-of-the-art results with significant improvement to the current technologies.
Deep Learning for 3D Point Clouds
As an efficient 3D vision solution, point clouds have been widely applied into diverse engineering scenarios, including immersive media communication, autonomous driving, reverse engineering, robots, topography mapping, digital twin city, medical analysis, digital museum, etc. Thanks to the great developments of deep learning theories and methods, 3D point cloud technologies have undergone fast growth during the past few years, including diverse processing and understanding tasks. Human and machine perception can be benefited from the success of using deep learning approaches, which can significantly improve 3D perception modeling and optimization, as well as 3D pre-trained and large models. This book delves into these research frontiers of deep learning-based point cloud technologies. The subject of this book focuses on diverse intelligent processing technologies for the fast-growing 3D point cloud applications, especially using deep learning-based approaches. The deep learning-based enhancement and analysis methods are elaborated in detail, as well as the pre-trained and large models with 3D point clouds. This book carefully presents and discusses the newest progresses in the field of deep learning-based point cloud technologies, including basic concepts, fundamental background knowledge, enhancement, analysis, 3D pre-trained and large models, multi-modal learning, open source projects, engineering applications, and future prospects. Readers can systematically learn the knowledge and the latest developments in the field of deep learning-based point cloud technologies. This book provides vivid illustrations and examples, and the intelligent processing methods for 3D point clouds. Readers can be equipped with an in-depth understanding of the latest advancements of this rapidly developing research field.
Point Cloud Compression
3D point clouds have broad applications across various industries and have contributed to advancements in fields such as autonomous driving, immersive media, metaverse, and cultural heritage protection. With the fast growth of 3D point cloud data and its applications, the need for efficient compression technologies has become paramount. This book delves into the forefront of point cloud compression, exploring key technologies, standardization efforts, and future prospects. This comprehensive book uncovers the foundational concepts, data acquisition methods, and datasets associated with point cloud compression. By examining the fundamental compression technologies, readers can obtain a clear understanding of prediction coding, transform coding, quantization techniques, and entropy coding. Through vivid illustrations and examples, the book elucidates how these techniques have evolved over the years and their potentials for the future. To provide a complete picture, the book presents cutting-edge research methods in point cloud compression and facilitates comparisons among them. Readers can be equipped with an in-depth understanding of the latest advancements, and can gain insights into the various approaches employed in this dynamic field. Another distinguishing aspect of this book is its exploration of standardization works for point cloud compression. Notable standards, such as MPEG G-PCC, AVS PCC, and MPEG V-PCC, are thoroughly illustrated. By delving into the methods used in geometry-based, video-based, and deep learning-based compression, readers become familiar with the latest breakthroughs in the standard communities.