17 dépôts
Neural network architectures for recognizing and categorizing 3D shapes.
Explore 17 awesome GitHub repositories matching part of an awesome list · 3D Object Classification. Refine with filters or upvote what's useful.
PointNet est une architecture de deep learning conçue pour traiter et classifier des nuages de points 3D bruts directement, sans voxelisation. Il fournit un système pour la classification d'objets 3D, des frameworks de segmentation sémantique pour partitionner les nuages en catégories, et des outils pour visualiser les formes 3D. Le projet utilise un réseau de transformation pour aligner les nuages de points dans un espace de coordonnées canonique et emploie une agrégation basée sur des fonctions symétriques pour condenser les caractéristiques point par point en vecteurs globaux, indépendamment de l'ordre des points. Il dispose également d'une architecture de regroupement multi-échelle pour extraire des caractéristiques géométriques hiérarchiques à différentes échelles spatiales. Le système inclut des pipelines de données pour convertir les données brutes d'analyse intérieure 3D et de forme en fichiers HDF5 pour l'entraînement. Les capacités d'évaluation couvrent les métriques de précision de segmentation et le rendu des nuages de points 3D mal classés en images à trois vues pour l'analyse d'erreurs.
Deep learning on point sets for classification and segmentation.
PointNet++ is a deep learning framework designed for processing and classifying 3D point cloud data. It utilizes a hierarchical feature learning architecture to extract geometric patterns from sampled 3D point sets. The framework implements a variety of 3D analysis tools, including a point cloud classifier for categorizing objects based on spatial coordinates and surface normals, a semantic scene segmenter for labeling surfaces in large-scale environments, and a tool for 3D object part segmentation. The system covers a broad range of capabilities including geometric feature extraction, 3D da
Deep hierarchical feature learning on point sets.
MeshCNN is a general-purpose deep neural network for 3D triangular meshes, which can be used for tasks such as 3D shape classification or segmentation. This framework includes convolution, pooling and unpooling layers which are applied directly on the mesh edges.
Neural network for processing 3D triangular meshes.
3D Generative Adversarial Network
Probabilistic latent space modeling for 3D shapes.
This repository contains the implementation of our papers related with O-CNN. The code is released under the MIT license.
Octree-based convolutional neural networks for shape analysis.
OctNet uses efficient space partitioning structures (i.e. octrees) to reduce memory and compute requirements of 3D convolutional neural networks, thereby enabling deep learning at high resolutions.
Octree-based deep 3D representation learning.
PointConv: Deep Convolutional Networks on 3D Point Clouds. CVPR 2019 Wenxuan Wu, Zhongang Qi, Li Fuxin.
Convolutional operator for point cloud processing.
Created by Itai Lang, Asaf Manor, and Shai Avidan from Tel Aviv University.
Differentiable point cloud sampling for deep learning.
3D/Volumetric Convolutional Neural Networks with Theano+Lasagne.
3D convolutional neural network for real-time object recognition.
Created by Yutong Feng, Yifan Feng, Haoxuan You, Xibin Zhao, Yue Gao from Tsinghua University.
Mesh-based neural network for shape representation.
Volumetric CNN (Convolutional Neural Networks) for Object Classification on 3D Data, with Torch implementation.
Volumetric and multi-view CNNs for 3D object classification.
Voxel-Based Variational Autoencoders, VAE GUI, and Convnets for Classification
Generative and discriminative voxel modeling with CNNs.
Here you can find the code for the BMVC 2017 version of "Orientation-boosted Voxel Nets for 3D Object Recognition", a.k.a ORION.
Orientation-boosted voxel nets for object recognition.
Paper: Feedback Networks, CVPR 2017.
Feedback-based neural network architecture.
SpiderCNN: Deep Learning on Point Sets with Parameterized Convolutional Filters. ECCV 2018 Yifan Xu, Tianqi Fan, Mingye Xu, Long Zeng, Yu Qiao.
Neural network architecture for point cloud classification.
1. Python (with necessary common libraries such as numpy, scipy, etc.) 2. TensorFlow 3. You need to prepare your data in *.mat file with the following format: - 'points': N x 3 array (x, y, z coordinates of the point cloud) - 'labels': N x 1 array (1-based integer per-point labels) - 'category':…
Deep network for 3D shape understanding.
Field Probing Neural Networks for 3D Data
Field probing neural networks for 3D data.