PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation (PAPER EXPLAINED)

Автор: Minimal Debug

Загружено: 2021-08-24

Просмотров: 25331

Описание:

GitHub: GitHub: https://github.com/aldipiroli/pointnet
Blog: https://minimal-debug.github.io/paper...
Arxiv: https://arxiv.org/abs/1612.00593

---
Abstract:
Point cloud is an important type of geometric data structure. Due to its irregular format, most researchers transform such data to regular 3D voxel grids or collections of images. This, however, renders data unnecessarily voluminous and causes issues. In this paper, we design a novel type of neural network that directly consumes point clouds and well respects the permutation invariance of points in the input. Our network, named PointNet, provides a unified architecture for applications ranging from object classification, part segmentation, to scene semantic parsing. Though simple, PointNet is highly efficient and effective. Empirically, it shows strong performance on par or even better than state of the art. Theoretically, we provide analysis towards understanding of what the network has learnt and why the network is robust with respect to input perturbation and corruption.

PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation (PAPER EXPLAINED)

Доступные форматы для скачивания:

Скачать видео mp4

Информация по загрузке:

Скачать аудио mp3

Похожие видео

VoxelNet: End-to-End Learning for Point Cloud Based 3D Object (PAPER EXPLAINED)

VoxelNet: End-to-End Learning for Point Cloud Based 3D Object (PAPER EXPLAINED)

PointNet | Лекция 43 (часть 1) | Прикладное глубокое обучение

PointNet | Лекция 43 (часть 1) | Прикладное глубокое обучение

PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation

PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation

Как происходит модернизация остаточных соединений [mHC]

Как происходит модернизация остаточных соединений [mHC]

LLM и GPT - как работают большие языковые модели? Визуальное введение в трансформеры

LLM и GPT - как работают большие языковые модели? Визуальное введение в трансформеры

2. How PointNet works as the pioneer of 3D point cloud backbone

2. How PointNet works as the pioneer of 3D point cloud backbone

Current Approaches and Future Directions for Point Cloud Object Detection in Intelligent Agents

Current Approaches and Future Directions for Point Cloud Object Detection in Intelligent Agents

[SGP-2022] Deep Learning on Point Clouds

[SGP-2022] Deep Learning on Point Clouds

Президент выводит войска? / Спецборт срочно вылетел в Москву

Президент выводит войска? / Спецборт срочно вылетел в Москву

3. How PointNet++ works on improving 3D point cloud backbone

3. How PointNet++ works on improving 3D point cloud backbone

Даулет Жангузин, Groq, Cohere, Lyft - Советы программистам от 10х инженера из Кремниевой Долины

Даулет Жангузин, Groq, Cohere, Lyft - Советы программистам от 10х инженера из Кремниевой Долины

[3D Point Cloud Data Processing] Capter 10. Overview of Deep Learning on Point-cloud

[3D Point Cloud Data Processing] Capter 10. Overview of Deep Learning on Point-cloud

Point Cloud Classification - Keras Code Examples

Point Cloud Classification - Keras Code Examples

Graph Neural Networks for Point Cloud Processing

Graph Neural Networks for Point Cloud Processing

PointNet++ | Lecture 43 (Part 2) | Applied Deep Learning

PointNet++ | Lecture 43 (Part 2) | Applied Deep Learning

Градиентный спуск, как обучаются нейросети | Глава 2, Глубинное обучение

Градиентный спуск, как обучаются нейросети | Глава 2, Глубинное обучение

Почему «Трансформеры» заменяют CNN?

Почему «Трансформеры» заменяют CNN?

Momentum Contrast for Unsupervised Visual Representation Learning - PAPER EXPLAINED

Momentum Contrast for Unsupervised Visual Representation Learning - PAPER EXPLAINED

3D Point Cloud Segmentation with SuperPoint Transformers (and Python)

3D Point Cloud Segmentation with SuperPoint Transformers (and Python)

Philipp Krähenbühl - Point-based object detection

Philipp Krähenbühl - Point-based object detection