A5091161586- Remote Sensing and LiDAR Applications
- Advanced Neural Network Applications
- Advanced Image and Video Retrieval Techniques
- Advanced Vision and Imaging
- 3D Surveying and Cultural Heritage
- Image Processing and 3D Reconstruction
Wuhan University
2022-2023
Taking depth into consideration has been proven to improve the performance of semantic segmentation through providing additional geometry information. Most existing works adopt a two-stream network, extracting features from color images and separately using two branches same structure, which suffer high memory computation costs. We find that acquired by simple downsampling can also play complementary part in task, sometimes even better than scheme with branches. In this paper, novel...
In recent years, exploring integrated heterogeneous features architecture has become one of the hot spots in 3D point cloud understanding. However, efficacy end-to-end training enhancing precision multi-view fusion for segmentation and its flexibility remain limited. Furthermore, it is worth highlighting that prior studies have consistently employed encoderdecoder architectures, predominantly emphasizing refinement encoder designs, while relatively neglecting significance decoders which...