Publications

2024

1. 

   Learning SO(3)-Invariant Semantic Correspondence via Local Shape Transform

   Chunghyun Park*, Seungwook Kim*, Jaesik Park, and Minsu Cho

   IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

   Abs arXiv

   Establishing accurate 3D correspondences between shapes stands as a pivotal challenge with profound implications for computer vision and robotics. However, existing self-supervised methods for this problem assume perfect input shape alignment, restricting their real-world applicability. In this work, we introduce a novel self-supervised Rotation-Invariant 3D correspondence learner with Local Shape Transform, dubbed RIST, that learns to establish dense correspondences between shapes even under challenging intra-class variations and arbitrary orientations. Specifically, RIST learns to dynamically formulate an SO(3)-invariant local shape transform for each point, which maps the SO(3)-equivariant global shape descriptor of the input shape to a local shape descriptor. These local shape descriptors are provided as inputs to our decoder to facilitate point cloud self- and cross-reconstruction. Our proposed self-supervised training pipeline encourages semantically corresponding points from different shapes to be mapped to similar local shape descriptors, enabling RIST to establish dense point-wise correspondences. RIST demonstrates state-of-the-art performances on 3D part label transfer and semantic keypoint transfer given arbitrarily rotated point cloud pairs, outperforming existing methods by significant margins.

2023

1. 

   Stable and Consistent Prediction of 3D Characteristic Orientation via Invariant Residual Learning

   Seungwook Kim*,  Chunghyun Park*, Yoonwoo Jeong, Jaesik Park, and Minsu Cho

   International Conference on Machine Learning (ICML), 2023

   Abs arXiv

   Learning to predict reliable characteristic orientations of 3D point clouds is an important yet challenging problem, as different point clouds of the same class may have largely varying appearances. In this work, we introduce a novel method to decouple the shape geometry and semantics of the input point cloud to achieve both stability and consistency. The proposed method integrates shape-geometry-based SO(3)-equivariant learning and shape-semantics-based SO(3)-invariant residual learning, where a final characteristic orientation is obtained by calibrating an SO(3)-equivariant orientation hypothesis using an SO(3)-invariant residual rotation. In experiments, the proposed method not only demonstrates superior stability and consistency but also exhibits state-of-the-art performances when applied to point cloud part segmentation, given randomly rotated inputs.

2022

1. 

   SeLCA: Self-Supervised Learning of Canonical Axis

   Seungwook Kim*, Yoonwoo Jeong*,  Chunghyun Park*, Jaesik Park, and Minsu Cho

   NeurIPS Workshop on Symmetry and Geometry in Neural Representations, 2022

   Abs PDF

   Robustness to rotation is critical for point cloud understanding tasks, as point cloud features can be affected dramatically with respect to prevalent rotation changes. In this work, we propose SeLCA, a novel self-supervised learning framework to learn to the canonical axis of point clouds in a probabilistic manner. In essence, we propose to learn rotationalequivariance by predicting the canonical axis of point clouds, and achieve rotationalinvariance by aligning the point clouds using their predicted canonical axis. When integrated into a rotation-sensitive pipeline, SeLCA achieves competitive performances on the ModelNet40 classification task under unseen rotations. Most interestingly, our proposed method also shows high robustness to various real-world point cloud corruptions presented by the ModelNet40-C dataset, compared to the state-of-the-art rotation-invariant method.

2. 

   PointMixer: MLP-Mixer for Point Cloud Understanding

   Jaesung Choe*,  Chunghyun Park*, Francois Rameau, Jaesik Park, and In So Kweon

   European Conference on Computer Vision (ECCV), 2022

   Abs arXiv Code

   MLP-Mixer has newly appeared as a new challenger against the realm of CNNs and Transformer. Despite its simplicity compared to Transformer, the concept of channel-mixing MLPs and token-mixing MLPs achieves noticeable performance in image recognition tasks. Unlike images, point clouds are inherently sparse, unordered and irregular, which limits the direct use of MLP-Mixer for point cloud understanding. To overcome these limitations, we propose PointMixer, a universal point set operator that facilitates information sharing among unstructured 3D point cloud. By simply replacing token-mixing MLPs with Softmax function, PointMixer can “mix" features within/between point sets. By doing so, PointMixer can be broadly used for intra-set, inter-set, and hierarchical-set mixing. We demonstrate that various channel-wise feature aggregation in numerous point sets is better than self-attention layers or dense token-wise interaction in a view of parameter efficiency and accuracy. Extensive experiments show the competitive or superior performance of PointMixer in semantic segmentation, classification, and reconstruction against Transformer-based methods.

3. 

   Fast Point Transformer

   Chunghyun Park, Yoonwoo Jeong, Minsu Cho, and Jaesik Park

   IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022

   Silver Prize @ Samsung HumanTech Paper Award
   Qualcomm Innovation Fellowship Winner

   Abs arXiv Code Website

   The recent success of neural networks enables a better interpretation of 3D point clouds, but processing a large-scale 3D scene remains a challenging problem. Most current approaches divide a large-scale scene into small regions and combine the local predictions together. However, this scheme inevitably involves additional stages for pre- and post-processing and may also degrade the final output due to predictions in a local perspective. This paper introduces Fast Point Transformer that consists of a new lightweight self-attention layer. Our approach encodes continuous 3D coordinates, and the voxel hashing-based architecture boosts computational efficiency. The proposed method is demonstrated with 3D semantic segmentation and 3D detection. The accuracy of our approach is competitive to the best voxel-based method, and our network achieves 129 times faster inference time than the state-of-the-art, Point Transformer, with a reasonable accuracy trade-off in 3D semantic segmentation on S3DIS dataset.

2021

1. 

   Improved Classification and Localization Approach to Small Bowel Capsule Endoscopy using Convolutional Neural Network

   Yunseob Hwang*, Han Hee Lee*,  Chunghyun Park, Bayu Adhi Tama, Jin Su Kim, Dae Young Cheung, Woo Chul Chung, Young-Seok Cho, Kang-Moon Lee, Myung-Gyu Choi, and Seungchul Lee

   Digestive Endoscopy, 2021

   Abs HTML PDF

   Although great advances in artificial intelligence for interpreting small bowel capsule endoscopy (SBCE) images have been made in recent years, its practical use is still limited. The aim of this study was to develop a more practical convolutional neural network (CNN) algorithm for the automatic detection of various small bowel lesions.