google/draco
Draco
Draco is a library and toolset for compressing, transcoding, and decoding 3D geometric meshes and point cloud data. Its primary purpose is to reduce storage size and transmission bandwidth for 3D assets.
The project includes a geometry optimizer specifically for glTF file containers to reduce asset footprints. It also features a hardened decoder designed to process malformed or untrusted 3D geometric data safely to prevent memory corruption and crashes.
The software covers a broad range of 3D data processing capabilities, including geometric data reconstruction, point attribute management, and the embedding of custom hierarchical metadata. It supports runtime decoding for real-time rendering and provides workflows for transcoding assets between different formats.
The project supports cross-platform compilation for various target architectures and operating systems.
Features
- 3D Mesh Compression - Reduces the file size of 3D geometric meshes to lower storage costs and speed up network transmission.
- Mesh Encoders - Encodes triangular meshes using sequential or edgebreaker methods to optimize the storage of vertices and faces.
- Point Cloud Compressors - Encodes large sets of 3D point data into a compact binary format for efficient network transfer.
- Point Cloud Compression - Reduces the size of point cloud data by encoding geometric attributes into a compact binary format.
- Data Compression Algorithms - Implements specialized compression algorithms to reduce the storage footprint of 3D meshes and point clouds.
- glTF Geometry Compression - Integrates geometry compression into glTF file containers to reduce asset footprints without changing the file format.
- Runtime Asset Decoding - Decompresses geometric meshes and point clouds during execution to render optimized assets in real time.
- 3D Asset Pipelines - Integrates geometry compression into glTF containers to reduce the footprint of 3D models for web and mobile.
- 3D Model Format Converters - Provides tools for translating 3D geometry between different industry-standard file formats.
- Geometric Data Compression - Encodes meshes and point clouds to reduce storage size and transmission bandwidth.
- Geometric Data Decompression - Restores compressed files back into standard formats for use in 3D applications.
- Geometry Optimizers - Optimizes 3D geometry specifically for integration into glTF file containers to reduce asset footprints.
- Point Cloud Compression - Encodes large sets of 3D point data into compact binary formats for more efficient storage and sharing.
- In-Memory Decompression Runtimes - Decodes compressed geometric data during execution to expand payloads directly into process memory for dynamic rendering.
- 3D Data Processing Libraries - Provides a comprehensive library for compressing and decompressing 3D geometric meshes and point clouds.
- Hardened Mesh Decoders - Provides a secure decompression implementation designed to process malformed or untrusted 3D geometric data safely.
- Secure Decoders - Processes untrusted or malformed 3D geometric files using hardened decoders to prevent memory corruption.
- Hardened Data Decoders - Processes malformed or malicious geometric meshes safely using a hardened decoder to prevent crashes and memory corruption.
- Memory Bounds Checking - Implements strict bounds checking and validation to prevent memory corruption from malformed 3D geometric data.
- Point Cloud Reconstruction - Reconstructs 3D surfaces, point clouds, and triangular meshes from compressed buffers.
- Binary Serialization - Stores compressed geometric data in a flat binary layout for fast sequential reading and minimal memory overhead.
- Data Transcoders - Converts 3D meshes and point clouds between different compressed and standard geometric formats.
- Coordinate-Specific Scalar Quantization - Maps high-precision floating point coordinates to a smaller integer grid to reduce the number of bits per vertex.
- Edgebreaker Mesh Encoding - Encodes triangular meshes by decomposing the connectivity into a sequence of edges to eliminate redundant vertex indices.
- Predictive Geometry Encoding - Reduces attribute size by storing the difference between predicted and actual values using quantization.
- Real-Time 3D Rendering Engines - Supports runtime decompression of 3D assets to allow complex models to be displayed in real-time applications.
- Data Compression and Decompression - Restores compressed geometric data into its original 3D representation for rendering.
- Data Encoders and Decoders - Translates encoded geometric meshes and point clouds back into usable formats using specialized decoders.
- Attribute-Specific Bitstream Packing - Organizes vertex normals, colors, and positions into separate bit-packed streams to optimize compression.
- Compressed Asset Importers - Converts compressed files back into standard mesh formats during the design phase for use in development environments.
- 点云处理 - Library for compressing 3D geometric meshes and point clouds.
