Closed ghost closed 3 years ago
Few thoughts:
Few thoughts:
- COCO has larger range of objects, so FPN is important. Current Panoptic-DeepLab only uses a single scale feature for detection. I would assume using FPN can improve performance a lot on COCO.
- AP metric itself is in favor of high recall. This is especially the case for COCO dataset. That's why top-down methods like Mask R-CNN shine on COCO.
- I did not find the optimal training parameters on COCO.
Thank you so much!
Hi, I am curious why the Instance segmentation performance on Cityscapes seems acceptable, but that on COCO is so poor?
What do you think may be the reason for this? Thank you in advance.