[ICCV, 2021] Cloud Transformers: A Universal Approach To Point Cloud Processing Tasks

Related tags

Computer Vision cloud_transformers

Overview

Cloud Transformers: A Universal Approach To Point Cloud Processing Tasks

This is an official PyTorch code repository of the paper "Cloud Transformers: A Universal Approach To Point Cloud Processing Tasks " (ICCV, 2021).

Here, we present a versatile point cloud processing block that yields state-of-the-art results on many tasks.
The key idea is to process point clouds with many cheap low-dimensional different projections followed by standard convolutions. And we do so both in parallel and sequentially.

Datasets

We provide links to the datasets we used to train/evaluate. After unpacking and preparation, please edit the dataset path (data:path field) in configs/*.yaml

Pre-trained models

We provide our pre-trained models' weights in a single archive.

Building Dependencies

To install and build all the modules required, please run:

bash ./install_deps.sh

Code Structure

In layers/cloud_transform.py the core operations are implemented (rasterization Splat and de-rasterization Slice). While in layers\mutihead_ct_*.py we provide slightly different versions of Multi-Headed Cloud Transform (MHCT).

The model zoo is situated in model_zoo, where the models for corresponding tasks are constructed of Multi-Headed Cloud Transforms.

Run

We train our models in multi-GPU setting using DistributedDataParallel. To train on n GPUs, please run the following commands:

python train_${SCRIPT_NAME}.py ${EXP_NAME} -c configs/${CONFIG_NAME}.yaml --master localhost:3315 --rank 0 --num_nodes n
...
python train_${SCRIPT_NAME}.py ${EXP_NAME} -c configs/${CONFIG_NAME}.yaml --master localhost:3315 --rank  --num_nodes n

The semantics for evaluation scripts is almost the same:

python eval_${SCRIPT_NAME}.py ${EXP_NAME} -c configs/eval/${CONFIG_NAME}.yaml

Cite

If you find our work helpful, please do not hesitate to cite us.

@inproceedings{mazur2021cloudtransformers,
  title={Cloud Transformers: A Universal Approach To Point Cloud Processing Tasks},
  author={Mazur, Kirill and Lempitsky, Victor},
  booktitle={International Conference on Computer Vision (ICCV)},
  year={2021}
}

Open Source research tool to search, browse, analyze and explore large document collections by Semantic Search Engine and Open Source Text Mining & Text Analytics platform (Integrates ETL for document processing, OCR for images & PDF, named entity recognition for persons, organizations & locations, metadata management by thesaurus & ontologies, search user interface & search apps for fulltext search, faceted search & knowledge graph)

Open Semantic Search https://opensemanticsearch.org Integrated search server, ETL framework for document processing (crawling, text extraction, text a

684 Jan 6, 2023

Comments

ScanObjectNN using 2k points?

Hi Team, Thank you very much for releasing the code.

Congratulations on state-of-the-art results on the ScanObjectNN dataset.

Can you please confirm if the results in Table 1 for CT are using 2048 input points instead of 1024 points?

Thanks in Advance.

opened by sheshap 0
Visualization

Dear authors,

Thanks for this excellent work.

Can you please provide any suggestions to visualize the segmentation results (e.g., Figure.1 Top Left)? I am grateful for your help.

Best Wishes, Haoran

opened by haoranD 0

[ICCV, 2021] Cloud Transformers: A Universal Approach To Point Cloud Processing Tasks

Related tags

Overview

Cloud Transformers: A Universal Approach To Point Cloud Processing Tasks

Datasets

Pre-trained models

Building Dependencies

Code Structure

Run

Cite

You might also like...

Implement 'Single Shot Text Detector with Regional Attention, ICCV 2017 Spotlight'

Image processing in Python

A post-processing tool for scanned sheets of paper.

Detect handwritten words in a text-line (classic image processing method).

Generic framework for historical document processing

A tool for extracting text from scanned documents (via OCR), with user-defined post-processing.

Solution for Problem 1 by team codesquad for AIDL 2020. Uses ML Kit for OCR and OpenCV for image processing

scantailor - Scan Tailor is an interactive post-processing tool for scanned pages.

Comments

ScanObjectNN using 2k points?

Visualization

Owner

Visual Understanding Lab @ Samsung AI Center Moscow

(CVPR 2021) Back-tracing Representative Points for Voting-based 3D Object Detection in Point Clouds

code for our ICCV 2021 paper "DeepCAD: A Deep Generative Network for Computer-Aided Design Models"

Dataset and Code for ICCV 2021 paper "Real-world Video Super-resolution: A Benchmark Dataset and A Decomposition based Learning Scheme"

An official PyTorch implementation of the paper "Learning by Aligning: Visible-Infrared Person Re-identification using Cross-Modal Correspondences", ICCV 2021.

A selectional auto-encoder approach for document image binarization

Code for the paper: Fusformer: A Transformer-based Fusion Approach for Hyperspectral Image Super-resolution

A python script based on opencv and paddleocr, which can automatically pick up tasks, make cookies, and receive rewards in the Destiny 2 Dawning Oven

pyntcloud is a Python library for working with 3D point clouds.

Code for CVPR 2022 paper "SoftGroup for Instance Segmentation on 3D Point Clouds"

Code for the paper "DewarpNet: Single-Image Document Unwarping With Stacked 3D and 2D Regression Networks" (ICCV '19)