computer-vision-7

Few-Shot 3D Point Cloud Classification

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in few-shot-3d-point-cloud-classification-7

Trend

Dataset

Best Model

Actions

ModelNet40 10-way (10-shot)

ModelNet40 10-way (20-shot)

Libraries

i

Use these libraries to find few-shot-3d-point-cloud-classification-7 models and implementations

runpeidong/act

3 papers 89

Datasets

ModelNet

Subtasks

No subtasks available.

Most implemented papers

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

C. Qi, Hao Su, Kaichun Mo, L. Guibas•Thu Dec 01 2016

This paper designs a novel type of neural network that directly consumes point clouds, which well respects the permutation invariance of points in the input and provides a unified architecture for applications ranging from object classification, part segmentation, to scene semantic parsing.

16162

Content

ModelNet40 5-way (10-shot)

ModelNet40 5-way (20-shot)

ScanObjectNN 10-way (10-shot)

ScanObjectNN 10-way (20-shot)

ScanObjectNN 5-way (10-shot)

ScanObjectNN 5-way (20-shot)

yangyanli/PointCNN

2 papers 1,349

vinits5/learning3d

2 papers 646

lmd0311/pointmamba

2 papers 231

0

Paper Graph

PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space

C. Qi, Hao Su, L. Guibas, L. Yi•Tue Jun 06 2017

A hierarchical neural network that applies PointNet recursively on a nested partitioning of the input point set and proposes novel set learning layers to adaptively combine features from multiple scales to learn deep point set features efficiently and robustly.

12754 0

Paper Graph

Attention is All you Need

Noam Shazeer, Ashish Vaswani, Lukasz Kaiser, Jakob Uszkoreit, Niki Parmar, I. Polosukhin, Llion Jones, Aidan N. Gomez•Sun Jun 11 2017

A new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely is proposed, which generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.

164803 0

Paper Graph

Dynamic Graph CNN for Learning on Point Clouds

M. Bronstein, Ziwei Liu, Yue Wang, Yongbin Sun, S. Sarma, J. Solomon•Tue Jan 23 2018

This work proposes a new neural network module suitable for CNN-based high-level tasks on point clouds, including classification and segmentation called EdgeConv, which acts on graphs dynamically computed in each layer of the network.

6827 0

Paper Graph

Masked Autoencoders for Point Cloud Self-supervised Learning

W. Liu, Yonghong Tian, Yatian Pang, Francis E. H. Tay, Wenxiao Wang, Liuliang Yuan•Sat Mar 12 2022

A simple architecture entirely based on standard Transformers can surpass dedicated Transformer models from supervised learning and inspires the feasibility of applying unified architectures from languages and images to the point cloud.

634 0

Paper Graph

Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?

Zekun Qi, Runpei Dong, Zheng Ge, Kaisheng Ma, Li Yi, Linfeng Zhang, Junbo Zhang, Jian‐Yuan Sun•Thu Dec 15 2022

This paper revisits masked modeling in a unified fashion of knowledge distillation, and shows that foundational Transformers pretrained with 2D images or natural languages can help self-supervised 3D representation learning through training Autoencoders as Cross-Modal Teachers (ACT).

139 0

Paper Graph

Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-training

Hongsheng Li, Renrui Zhang, Peng Gao, Ziyu Guo, Rongyao Fang, Bingyan Zhao, Dong Wang, Y. Qiao•Fri May 27 2022

Point-M2AE is proposed, a strong Multi-scale MAE pre-training framework for hierarchical self-supervised learning of 3D point clouds that modifications the encoder and decoder into pyramid architectures to progressively model spatial geometries and capture both fine-grained and high-level semantics of3D shapes.

354 0

Paper Graph

Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining

Zekun Qi, Runpei Dong, Guo Fan, Zheng Ge, Xiangyu Zhang, Kaisheng Ma, Li Yi•Sat Feb 04 2023

ReCon is trained to learn from both generative modeling teachers and single/cross-modal contrastive teachers through ensemble distillation, where the generative student guides the contrastive student.

192 0

Paper Graph

Instance-aware Dynamic Prompt Tuning for Pre-trained Point Cloud Models

Shutao Xia, Yaohua Zha, Jinpeng Wang, Tao Dai, Bin Chen, Zhi Wang•Thu Apr 13 2023

The essence of IDPT is to develop a dynamic prompt generation module to perceive semantic prior features of each point cloud instance and generate adaptive prompt tokens to enhance the model's robustness, providing a promising solution to parameter-efficient learning for pre-trained point cloud models.

74 0

Paper Graph

Adding a benchmark result helps the community track progress.