Panoptic SwiftNet: Pyramidal Fusion for Real-time Panoptic Segmentation

TL;DR

Panoptic SwiftNet introduces an efficient pyramidal fusion approach for real-time panoptic segmentation, balancing backbone capacity and multi-scale features, suitable for large-scale remote sensing applications.

Contribution

It proposes a novel pyramidal fusion method with boundary-aware learning, enabling fast, accurate panoptic segmentation on high-resolution images with limited computational resources.

Findings

Outperforms state-of-the-art on BSB-Aerial dataset

Processes over 100 1MPx images per second on RTX3090

Effective for remote sensing imagery

Abstract

Dense panoptic prediction is a key ingredient in many existing applications such as autonomous driving, automated warehouses or remote sensing. Many of these applications require fast inference over large input resolutions on affordable or even embedded hardware. We propose to achieve this goal by trading off backbone capacity for multi-scale feature extraction. In comparison with contemporaneous approaches to panoptic segmentation, the main novelties of our method are efficient scale-equivariant feature extraction, cross-scale upsampling through pyramidal fusion and boundary-aware learning of pixel-to-instance assignment. The proposed method is very well suited for remote sensing imagery due to the huge number of pixels in typical city-wide and region-wide datasets. We present panoptic experiments on Cityscapes, Vistas, COCO and the BSB-Aerial dataset. Our models outperform the state of the art on the BSB-Aerial dataset while being able to process more than a hundred 1MPx images per second on a RTX3090 GPU with FP16 precision and TensorRT optimization.