Segment scene fast by change multi-path blocks in training to single-path when infer with RDRNet
Segment scene fast by change multi-path blocks in training to single-path when infer with RDRNet
Reparameterizable Dual-Resolution Network for Real-time Semantic Segmentation
arXiv paper abstract https://arxiv.org/abs/2406.12496
arXiv PDF paper https://arxiv.org/pdf/2406.12496
GitHub https://github.com/gyyang23/RDRNet
Semantic segmentation plays a key role in applications such as autonomous driving and medical image.
Although existing real-time semantic segmentation models achieve a commendable balance between accuracy and speed, their multi-path blocks still affect overall speed.
To address this issue, this study proposes a Reparameterizable Dual-Resolution Network (RDRNet) dedicated to real-time semantic segmentation.
… RDRNet employs a two-branch … utilizing multi-path blocks during training and reparameterizing … into single-path … during inference … enhancing … accuracy and inference speed
… propose the Reparameterizable Pyramid Pooling Module (RPPM) to enhance the feature representation of the pyramid pooling module without increasing its inference time.
… demonstrate that RDRNet outperforms existing state-of-the-art models in terms of both performance and speed …
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website