edecoux

aliases	tags	title	created
		Distributed and Decentralized Computing	Sep 06, 2022

• Distributed machine learning nodes train models independently.
• The central server transmits the initial model to the distributed nodes (devices).
• The nodes have a local model with their own data.

edecoux

Serverless Fabric

The serverless fabric combines the Function as a Service (FaaS) paradigm with serverless computing, hardware acceleration (GPU, FPGA and CPU), and a cross layer, multiconnectivity-enabled disaggregated Radio Access Network (RAN) into a single platform allowing over-the-top providers to fully use the 6G capabilities though well established cloud-native paradigms to develop and run applications. Each Network Function (NF) exposes a set of services.

• FaaS. Function as a service (FaaS) is a paradigm in which the customers are able to develop, run, and manage application functionalities without the trouble of building and maintaining the infrastructure.

• BaaS. Backend as a Service (BaaS) is an online service that handles a specific task over the cloud, such as authentication or notification. Both BaaS and FaaS require no resource management from the customers. While FaaS only offers to execute users’ functions, BaaS offers a complete online service.

Network Slicing

Network

edecoux

[[Towards Collaborative and Dynamic Spectrum Sharing via Interpretation of Spectrum Access Policies.pdf]]

---

Cognitive Radio

A number of definitions of a cognitive radio (CR) have been published in the literature. We are using here a definition that is provided by the IEEE P1900.1 standard:

• A type of radio in which communication systems are aware of their environment and internal state and can make decisions about their radio operating behavior based on that information and predefined objectives.

• Uses software-defined radio, adaptive radio, and other technologies to adjust automatically its behavior or operations to achieve desired objectives.”

edecoux

Visual SLAM

Visual SLAM algorithms: a survey from 2010 to 2016 https://share.summari.com/articles101186s41074-017-0027-2?utm_source=Webapp

• Technique for obtaining the 3D structure of an unknown environment.
• Originally proposed to achieve autonomous control of robots in robotics.
• Input of such SLAM is visual information only.

Elements of vSLAM

• Define a certain coordinate system for camera pose estimation and 3D reconstruction in an unknown environment. In the initialization, the global coordinate system should first be defined, and a part of the environment is reconstructed as an initial map. Tracking and mapping are performed to continuously estimate camera poses
• 2D-3D correspondences between the image and the map are first obtained from feature matching or feature tracking in the image. Camera pose is computed from the correspondences by solving the Perspective-n-Point (PnP) problem

edecoux

Visual SLAM

Visual SLAM algorithms: a survey from 2010 to 2016 https://share.summari.com/articles101186s41074-017-0027-2?utm_source=Webapp

• Simultaneous Localization and Mapping (SLAM) is a technique for obtaining the 3D structure of an unknown environment
• This technique was originally proposed to achieve autonomous control of robots in robotics.
• Since the input of such SLAM is visual information only, the technique is specifically referred to as visual SLAM (vSLAM).

Elements of vSLAM

• Define a certain coordinate system for camera pose estimation and 3D reconstruction in an unknown environment
• In the initialization, the global coordinate system should first be defined, and a part of the environment is reconstructed as an initial map
• Tracking and mapping are performed to continuously estimate camera poses

edecoux

Nanite Virtualized Geometry

https://docs.unrealengine.com/5.0/en-US/nanite-virtualized-geometry-in-unreal-engine/

• Multiple orders of magnitude increase in geometry complexity, higher triangle and objects counts than has been possible before in real-time
• Frame budgets are no longer constrained by polycounts, draw calls, and mesh memory usage
• Now possible to directly import film-quality source arts, such as ZBrush sculpts and photogrammetry scans
• Use high-poly detailing rather than baking detail into normal map textures
• Level of Detail (LOD) is automatically handled and no longer requires manual setup for individual mesh's LODs

edecoux

Photorealistic Rendering

The ray tracing algorithm is based on following the path of a ray of light through a scene as it interacts with and bounces off geometry in an environment.

Cameras

• A camera model determines how and from where the scene is being viewed. The most important function of the camera is to define the portion of the scene that will be recorded
• Pinhole cameras are the simplest form of camera. Connecting the pinhole to the edges of the film creates a double pyramid that extends into the scene. Objects that are not inside the pyramid cannot be imaged on the film, so the viewing volume is the same as before.

edecoux

Virtual Texturing

• enables you to create and use large-sized textures for a lower—and more consistent—memory footprint at runtime.
• Unreal Engine supports two virtual texturing methods: Runtime Virtual Texturing (RVT) and Streaming Virtual Texturing (SVT).
• RVT supplies an efficient way to render complex, procedurally generated, or layered materials.
• Streaming Virtual Texturing reduces texture memory overhead when using larger texture sizes, including support for virtual texture lightmaps and UDIM (U-Dimension). Streaming virtual textures is an alternative way to stream textures from disk compared to existing mip-based texture streaming.
• Enabling support for virtual texture lightmaps can improve streaming performance and quality of lightmap bakes.

Runtime Virtual Textures

Supports larger texture resolutions. Texel data cached in memory on demand. Texel data generated by the GPU at runtime. Well suited for texture data that can be rendered on demand, such as procedural textures or composited layere

edecoux

High-performance adaptive texture streaming and rendering of large 3D cities

• A high-performance texture streaming system that automatically adjusts the streaming workload at runtime based on measured frame latencies
• Proposes a novel texture streaming algorithm that automatically adjusts texture streaming rate based on measured frame latencies
• A mipmap color blending method to minimize texture pop-in artifacts and preexisting mipmapping artifacts
• An optional mesh clustering scheme to generate texture atlases by grouping nearby and similarly oriented meshes together

City rendering in GIS

edecoux

UE5

Nanite geometry

Virtual Texturing

Materials

• Materials are constructed through a network of visual scripting nodes (called Material Expressions) within the Material Editor