Code for the paper "Benchmarking and Analyzing Point Cloud Classification under Corruptions"

Last update: Dec 28, 2022

Overview

ModelNet-C

Code for the paper "Benchmarking and Analyzing Point Cloud Classification under Corruptions". For the latest updates, see: sites.google.com/view/modelnetc/home

Benchmarking and Analyzing Point Cloud Classification under Corruptions
Jiawei Ren, Liang Pan, Ziwei Liu

arXiv 2022

ModelNet-C [Download Link]

Get Started

Step 0. Clone the Repo

git clone https://github.com/jiawei-ren/ModelNet-C.git
cd ModelNet-C

Step 1. Set Up the Environment

Set up the environment by:

conda create --name modelnetc python=3.7.5
conda activate modelnetc
pip install -r requirements.txt
cd SimpleView/pointnet2_pyt && pip install -e . && cd -
pip install -e modelnetc_utils

Step 2. Prepare Data

Download ModelNet-C by:

cd data
gdown https://drive.google.com/uc?id=1KE6MmXMtfu_mgxg4qLPdEwVD5As8B0rm
unzip modelnet_c.zip && cd ..

Alternatively, you may download ModelNet40-C manually and extract it under data.

Step 3. Download Pretrained Models

Download pretrained models by

gdown https://drive.google.com/uc?id=11RONLZGg0ezxC16n57PiEZouqC5L0b_h
unzip pretrained_models.zip

Alternatively, you may download pretrained models manually and extract it under root directory.

Benchmark on ModelNet-C

Evaluation Commands

Evaluation commands are provided in EVALUATE.md.

Benchmark Results

Method	Reference	Standalone	mCE	Clean OA
DGCNN	Wang et al.	Yes	1.000	0.926
PointNet	Qi et al.	Yes	1.422	0.907
PointNet++	Qi et al.	Yes	1.072	0.930
RSCNN	Liu et al.	Yes	1.130	0.923
SimpleView	Goyal et al.	Yes	1.047	0.939
GDANet	Xu et al.	Yes	0.892	0.934
CurveNet	Xiang et al.	Yes	0.927	0.938
PAConv	Xu et al.	Yes	1.104	0.936
PCT	Guo et al.	Yes	0.925	0.930
RPC	Ren et al.	Yes	0.863	0.930
DGCNN+PointWOLF	Kim et al.	No	0.814	0.926
DGCNN+RSMix	Lee et al.	No	0.745	0.930
DGCNN+WOLFMix	Ren et al.	No	0.590	0.932
GDANet+WOLFMix	Ren et al.	No	0.571	0.934

*Standalone indicates if the method is a standalone architecture or a combination with augmentation or pretrain.

Todos

PointMixup
OcCo
PointBERT

Cite ModelNet-C

@article{
    ren2022modelnetc,
    title={Benchmarking and Analyzing Point Cloud Classification under Corruptions},
    author={Jiawei Ren and Liang Pan and Ziwei Liu},
    journal={arXiv:2202.03377},
    year={2022},
}

Acknowledgement

This codebase heavily borrows codes from the following repositories:

Code for the paper "Benchmarking and Analyzing Point Cloud Classification under Corruptions"

Related tags

Overview

ModelNet-C

Get Started

Step 0. Clone the Repo

Step 1. Set Up the Environment

Step 2. Prepare Data

Step 3. Download Pretrained Models

Benchmark on ModelNet-C

Evaluation Commands

Benchmark Results

Todos

Cite ModelNet-C

Acknowledgement

Owner

Jiawei Ren

PyTorch implementation of DD3D: Is Pseudo-Lidar needed for Monocular 3D Object detection?

Tutel MoE: An Optimized Mixture-of-Experts Implementation

Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers

RobustART: Benchmarking Robustness on Architecture Design and Training Techniques

Code and model benchmarks for "SEVIR : A Storm Event Imagery Dataset for Deep Learning Applications in Radar and Satellite Meteorology"

Multi-Objective Loss Balancing for Physics-Informed Deep Learning

Continual learning with sketched Jacobian approximations

Neural Network Libraries

Code Release for Learning to Adapt to Evolving Domains

Towards Implicit Text-Guided 3D Shape Generation (CVPR2022)

Tiny Kinetics-400 for test

Independent and minimal implementations of some reinforcement learning algorithms using PyTorch (including PPO, A3C, A2C, ...).

DanceTrack: Multiple Object Tracking in Uniform Appearance and Diverse Motion

Python implementation of NARS (Non-Axiomatic-Reasoning-System)

Face Recognition plus identification simply and fast | Python

Neural network-based build time estimation for additive manufacturing

The repository forked from NVlabs uses our data. (Differentiable rasterization applied to 3D model simplification tasks)

Wenet STT Python

A new play-and-plug method of controlling an existing generative model with conditioning attributes and their compositions.

The implementation of the CVPR2021 paper "Structure-Aware Face Clustering on a Large-Scale Graph with 10^7 Nodes"