Daan de Geus, Panagiotis Meletis, Gijs Dubbelman
We present a single network method for panoptic segmentation. This method combines the predictions from a jointly trained semantic and instance segmentation network using heuristics. Joint training is the first step towards an end-to-end panoptic segmentation network and is faster and more memory efficient than training and predicting with two networks, as done in previous work. The architecture consists of a ResNet-50 feature extractor shared by the semantic segmentation and instance segmentation branch. For instance segmentation, a Mask R-CNN type of architecture is used, while the semantic segmentation branch is augmented with a Pyramid Pooling Module. Results for this method are submitted to the COCO and Mapillary Joint Recognition Challenge 2018. Our approach achieves a PQ score of 17.6 on the Mapillary Vistas validation set and 27.2 on the COCO test-dev set.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Semantic Segmentation | Mapillary val | PQ | 17.6 | JSIS-Net (ResNet-50) |
| Semantic Segmentation | COCO test-dev | PQ | 27.2 | JSIS-Net |
| Semantic Segmentation | COCO test-dev | PQst | 23.4 | JSIS-Net |
| Semantic Segmentation | COCO test-dev | PQth | 29.6 | JSIS-Net |
| 10-shot image generation | Mapillary val | PQ | 17.6 | JSIS-Net (ResNet-50) |
| 10-shot image generation | COCO test-dev | PQ | 27.2 | JSIS-Net |
| 10-shot image generation | COCO test-dev | PQst | 23.4 | JSIS-Net |
| 10-shot image generation | COCO test-dev | PQth | 29.6 | JSIS-Net |
| Panoptic Segmentation | Mapillary val | PQ | 17.6 | JSIS-Net (ResNet-50) |
| Panoptic Segmentation | COCO test-dev | PQ | 27.2 | JSIS-Net |
| Panoptic Segmentation | COCO test-dev | PQst | 23.4 | JSIS-Net |
| Panoptic Segmentation | COCO test-dev | PQth | 29.6 | JSIS-Net |