| Paper ID | IVMSP-11.6 |
| Paper Title |
MPDNet: A 3D Missing Part Detection Network Based on Point Cloud Segmentation |
| Authors |
Zhaoxin Fan, Renmin University of China, China; Hongyan Liu, Tsinghua University, China; Jun He, Min Zhang, Xiaoyong Du, Renmin University of China, China |
| Session | IVMSP-11: Image & Video Segmentation |
| Location | Gather.Town |
| Session Time: | Wednesday, 09 June, 14:00 - 14:45 |
| Presentation Time: | Wednesday, 09 June, 14:00 - 14:45 |
| Presentation |
Poster
|
| Topic |
Image, Video, and Multidimensional Signal Processing: [IVELI] Electronic Imaging |
| IEEE Xplore Open Preview |
Click here to view in IEEE Xplore |
| Virtual Presentation |
Click here to watch in the Virtual Conference |
| Abstract |
Utilizing computer vision technologies for machinery missing part detection has been a hot research topic recently. Most of existing methods take images as input and utilize 2D object detection pipelines for detecting fault regions. However, 2D models can’t handle the situation when occlusion exists. Therefore, we propose MPDNet, a model exploits 3D point cloud pairs as input for missing part detection. In MPDNet, the missing part detection problem is transformed into a binary segmentation problem. The key idea is that difference between two point clouds can be fully perceived if they share the same encoder. We firstly propose a shared encode and abnormal lift module to find and enlarge difference between the target point cloud to be diagnosed and its corresponding predefined source point cloud . Then an attention based decoder is proposed to segment the source point cloud into two clusters: points that are missing in the target point cloud and points that are preserved in the target point cloud. What's more, a point cloud construction module is proposed as an auxiliary task to help the shared encoder to learn more discriminative features. Experiments on both synthetic and real world datasets have demonstrated the effectiveness of MPDNet. |
