awesome-repositories.com
Blog
awesome-repositories.com

Découvrez les meilleurs dépôts open-source grâce à notre recherche par IA.

ExplorerRecherches sélectionnéesAlternatives open sourceLogiciels auto-hébergésBlogPlan du site
ProjetÀ proposNotre méthodologiePresseServeur MCP
Mentions légalesConfidentialitéConditions d'utilisation
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
·
bmild avatar

bmild/nerf

0
View on GitHub↗
10,902 stars·1,438 forks·Jupyter Notebook·MIT·5 vuestancik.com/nerf↗

Nerf

This project is a framework for neural radiance fields used to synthesize three-dimensional environments from sets of two-dimensional images and camera poses. It functions as a volumetric rendering engine and scene synthesizer that optimizes neural representations of spatial volumes to generate novel views of complex 3D scenes.

The system implements a coordinate encoding system that transforms spatial coordinates into high-dimensional space to capture high-frequency geometric details. It also includes a neural mesh extractor that converts trained radiance fields into triangle meshes via marching cubes for use in 3D software.

The project covers a broad range of capabilities, including differentiable volumetric rendering, camera pose estimation, and the import of scene datasets and pre-trained model weights. It handles the full pipeline from camera-to-world ray projection and volumetric ray marching to the final extraction of polygonal geometry.

Features

  • Radiance Field Training Pipelines - Provides a complete framework for optimizing neural networks to create high-fidelity 3D scene representations from 2D images.
  • Novel View Synthesis Engines - Generates high-quality images of a trained 3D scene from camera perspectives not present in the original training set.
  • Neural Radiance Field Synthesizers - Optimizes neural representations of spatial volumes from images and camera poses to synthesize 3D environments.
  • Neural Scene Optimizers - Optimizes a neural network to synthesize a 3D scene using a set of 2D images and camera poses.
  • Spatial Coordinate Encodings - Transforms spatial coordinates into high-dimensional vectors using sine and cosine functions to capture high-frequency geometric details.
  • Neural Rendering - Optimizes neural network weights by comparing rendered ray samples against ground-truth pixels using a differentiable rendering pipeline.
  • Photorealistic Scene Synthesis - Creates high-fidelity digital 3D representations of real-world environments using sets of 2D images.
  • Adaptive Ray Marchers - Estimates pixel colors and opacity by sampling points along rays using learned spatial structures.
  • Volumetric Rendering Engines - Implements a volumetric rendering engine that estimates color and density along camera rays to generate photorealistic images.
  • High-Frequency Coordinate Mappers - Transforms input coordinates into a higher-dimensional space using periodic functions to capture high-frequency geometric details.
  • Model Architecture - Defines a neural network architecture that maps 3D spatial coordinates and view directions to volumetric color and density.
  • Pose Estimation - Calculates the precise position and orientation of cameras used to capture a scene to ensure accurate spatial alignment.
  • Isosurface Extractors - Converts a learned continuous volumetric density field into a discrete polygonal triangle mesh via marching cubes.
  • Volumetric Mesh Extractors - Converts trained neural radiance fields into triangle meshes using marching cubes for use in 3D software.
  • 3D Intersection Ray Calculators - Implements the mathematical projection of 2D image pixels into 3D world-space rays based on camera parameters.
  • Volumetric Mesh Exporters - Converts neural volumetric data into polygonal triangle meshes for use in traditional 3D modeling software.
  • Volumetric Mesh Extraction - Provides a neural mesh extractor that converts trained radiance fields into triangle meshes using marching cubes.
  • Positional Encodings - Implements a coordinate encoding system that transforms spatial coordinates into high-dimensional space to capture fine geometric details.
  • 3D Computer Vision - Neural radiance fields for high-quality view synthesis.
  • 3D Reconstruction - Neural radiance fields for high-quality view synthesis.
  • Neural Radiance Field Implementations - Original reference implementation using the TensorFlow framework.
  • Neural Radiance Fields - Original reference implementation using the TensorFlow framework.
  • Neural Scene Representation - Foundational neural radiance field model for view synthesis.
  • Neural Rendering - Listed in the “Neural Rendering” section of the Ailia Models awesome list.

Historique des stars

Graphique de l'historique des stars pour bmild/nerfGraphique de l'historique des stars pour bmild/nerf

Recherche par IA

Explorez plus de dépôts awesome

Décrivez vos besoins en langage naturel — l'IA classe des milliers de projets open source sélectionnés par pertinence.

Start searching with AI

Alternatives open source à Nerf

Projets open source similaires, classés selon le nombre de fonctionnalités partagées avec Nerf.
  • yenchenlin/nerf-pytorchAvatar de yenchenlin

    yenchenlin/nerf-pytorch

    6,037Voir sur GitHub↗

    This project is a PyTorch implementation of a Neural Radiance Field framework. It serves as a 3D scene synthesizer and differentiable volumetric renderer used to train volumetric representations of scenes by predicting color and density for 3D spatial coordinates. The system enables novel view synthesis, allowing for the generation of new images of complex 3D scenes from previously unseen perspectives. It supports 3D scene reconstruction by processing 2D images and camera poses to build a digital volumetric representation of a physical space. The framework includes capabilities for 3D model

    Python
    Voir sur GitHub↗6,037
  • google-research/multinerfAvatar de google-research

    google-research/multinerf

    3,806Voir sur GitHub↗

    MultiNeRF is a 3D scene reconstruction suite and framework for training Neural Radiance Fields to synthesize novel views from sets of 2D images. It provides a system for generating new perspectives of a scene by optimizing a neural network based on images and camera poses. The toolkit includes research implementations such as Mip-NeRF 360 and Ref-NeRF for high-fidelity volumetric rendering. It features a structure-from-motion pipeline to calculate camera positions and orientations from image datasets to prepare data for training. The project covers a full workflow for volumetric rendering, i

    Pythonnerfneural-radiance-fields
    Voir sur GitHub↗3,806
  • facebookresearch/pytorch3dAvatar de facebookresearch

    facebookresearch/pytorch3d

    9,902Voir sur GitHub↗

    PyTorch3D is a 3D geometric deep learning library and mesh processing toolkit designed for learning from point clouds and complex 3D surface geometries. It provides a collection of reusable components and data structures for deep learning with 3D data, including a framework for training and evaluating neural radiance fields to enable photorealistic view synthesis. The project features a differentiable 3D renderer that converts meshes and point clouds into 2D images while allowing gradients to flow back into the geometry and textures. This enables 3D shape optimization, where mesh geometry, te

    Python
    Voir sur GitHub↗9,902
  • ashawkey/stable-dreamfusionAvatar de ashawkey

    ashawkey/stable-dreamfusion

    8,841Voir sur GitHub↗

    This project is a diffusion-based 3D generator and image-to-3D reconstruction system. It translates natural language descriptions or two-dimensional images into three-dimensional assets using neural radiance fields and diffusion models. The system utilizes score-distillation sampling and diffusion-based guidance to refine 3D shapes without requiring 3D training data. It includes specialized tools for transforming neural representations into exportable meshes with texture and material data, as well as a pipeline for iterative optimization of geometry and textures. The project covers a broad r

    Python
    Voir sur GitHub↗8,841
Voir les 30 alternatives à Nerf→

Questions fréquentes

Que fait bmild/nerf ?

This project is a framework for neural radiance fields used to synthesize three-dimensional environments from sets of two-dimensional images and camera poses. It functions as a volumetric rendering engine and scene synthesizer that optimizes neural representations of spatial volumes to generate novel views of complex 3D scenes.

Quelles sont les fonctionnalités principales de bmild/nerf ?

Les fonctionnalités principales de bmild/nerf sont : Radiance Field Training Pipelines, Novel View Synthesis Engines, Neural Radiance Field Synthesizers, Neural Scene Optimizers, Spatial Coordinate Encodings, Neural Rendering, Photorealistic Scene Synthesis, Adaptive Ray Marchers.

Quelles sont les alternatives open-source à bmild/nerf ?

Les alternatives open-source à bmild/nerf incluent : yenchenlin/nerf-pytorch — This project is a PyTorch implementation of a Neural Radiance Field framework. It serves as a 3D scene synthesizer and… google-research/multinerf — MultiNeRF is a 3D scene reconstruction suite and framework for training Neural Radiance Fields to synthesize novel… facebookresearch/pytorch3d — PyTorch3D is a 3D geometric deep learning library and mesh processing toolkit designed for learning from point clouds… ashawkey/stable-dreamfusion — This project is a diffusion-based 3D generator and image-to-3D reconstruction system. It translates natural language… nerfstudio-project/nerfstudio — Nerfstudio is a modular development framework for training, visualizing, and exporting three-dimensional scene… nerfies/nerfies.github.io — This project is a computer vision pipeline and volumetric rendering system used to transform photos and videos into…