# You Only Group Once: Efficient Point-Cloud Processing with Token Representation and Relation Inference Module

arXiv: https://arxiv.org/abs/2103.09975 · HTML: https://mlsanity.com/paper/2103.09975/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Mar 2021.

## Built on by (most cited first, 13 of 13)

- [SPViT: Enabling Faster Vision Transformers via Soft Token Pruning](https://mlsanity.com/paper/2112.13890/) (2021)
- [IRISformer: Dense Vision Transformers for Single-Image Inverse Rendering in Indoor Scenes](https://mlsanity.com/paper/2206.08423/) (2022)
- [Image2Point: 3D Point-Cloud Understanding with 2D Image Pretrained Models](https://mlsanity.com/paper/2106.04180/) (2021)
- [Open-Vocabulary 3D Detection via Image-level Class and Debiased Cross-modal Contrastive Learning](https://mlsanity.com/paper/2207.01987/) (2022)
- [LCPFormer: Towards Effective 3D Point Cloud Analysis via Local Context Propagation in Transformers](https://mlsanity.com/paper/2210.12755/) (2022)
- [DetMatch: Two Teachers are Better Than One for Joint 2D and 3D Semi-Supervised Object Detection](https://mlsanity.com/paper/2203.09510/) (2022)
- [DELFlow: Dense Efficient Learning of Scene Flow for Large-Scale Point Clouds](https://mlsanity.com/paper/2308.04383/) (2023)
- [Collect-and-Distribute Transformer for 3D Point Cloud Analysis](https://mlsanity.com/paper/2306.01257/) (2023)
- [ESCAPE: Equivariant Shape Completion via Anchor Point Encoding](https://mlsanity.com/paper/2412.00952/) (2024)
- [PointHR: Exploring High-Resolution Architectures for 3D Point Cloud Segmentation](https://mlsanity.com/paper/2310.07743/) (2023)
- [CloudAttention: Efficient Multi-Scale Attention Scheme For 3D Point Cloud Learning](https://mlsanity.com/paper/2208.00524/) (2022)
- [Rethinking Attention Module Design for Point Cloud Analysis](https://mlsanity.com/paper/2407.19294/) (2024)
- [ESP-Zero: Unsupervised enhancement of zero-shot classification for Extremely Sparse Point cloud](https://mlsanity.com/paper/2404.19639/) (2024)

## Abstract (arXiv)

3D point-cloud-based perception is a challenging but crucial computer vision task. A point-cloud consists of a sparse, unstructured, and unordered set of points. To understand a point-cloud, previous point-based methods, such as PointNet++, extract visual features through hierarchically aggregation of local features. However, such methods have several critical limitations: 1) Such methods require several sampling and grouping operations, which slow down the inference speed. 2) Such methods spend an equal amount of computation on each points in a point-cloud, though many of points are redundant. 3) Such methods aggregate local features together through downsampling, which leads to information loss and hurts the perception performance. To overcome these challenges, we propose a novel, simple, and elegant deep learning model called YOGO (You Only Group Once). Compared with previous methods, YOGO only needs to sample and group a point-cloud once, so it is very efficient. Instead of operating on points, YOGO operates on a small number of tokens, each of which summarizes the point features in a sub-region. This allows us to avoid computing on the redundant points and thus boosts efficiency.Moreover, YOGO preserves point-wise features by projecting token features to point features although the computation is performed on tokens. This avoids information loss and can improve point-wise perception performance. We conduct thorough experiments to demonstrate that YOGO achieves at least 3.0x speedup over point-based baselines while delivering competitive classification and segmentation performance on the ModelNet, ShapeNetParts and S3DIS datasets.