| Full name | Efficient Multi-Task RGB-D Scene Analysis for Indoor Environments |
| Description | EMSANet is a lightweight neural network that enables real-time panoptic segmentation on an NVIDIA Jetson AGX Xavier. It comprises a fused RGB-D encoder with two lightweight ResNet34-NBt1D-based backbones, a decoder for semantic segmentation, and a decoder for class-agnostic instance segmentation. The results of both decoders are merged to derive a panoptic segmentation. Note, this model was trained (after the mentioned publication) for "PanopticNDT: Efficient and Robust Panoptic Mapping" (IROS 2023).
Weights are publicly available at: https://github.com/TUI-NICR/panoptic-mapping |
| Publication title | EMSANet: Efficient Multi-Task RGB-D Scene Analysis for Indoor Environments |
| Publication authors | Seichter, Daniel and Fischedick, Söhnke and Köhler, Mona and Gross, Horst-Michael |
| Publication venue | IJCNN 2022 |
| Publication URL | https://arxiv.org/abs/2207.04526 |
| Input Data Types | Uses Color,Uses Geometry Uses 2D |
| Programming language(s) | PyTorch CUDA |
| Hardware | Nvidia Jetson AGX Xavier/Orin |
| Source code or download URL | https://github.com/TUI-NICR/EMSANet |
| Submission creation date | 5 Feb, 2023 |
| Last edited | 23 Jan, 2024 |