search

dvlab-research / PanopticFCN

Fully Convolutional Networks for Panoptic Segmentation (CVPR2021 Oral)
Apache License 2.0
391 stars 53 forks source link

Why it works well? #9

Closed CuberrChen closed 3 years ago

CuberrChen commented 3 years ago

Hi! The point of view in this paper is very new, but I don’t quite understand its working mechanism.Can you explain briefly？

CuberrChen commented 3 years ago

Or what is the difference between it and solov2?

yanwei-li commented 3 years ago

Hi! In general, Panoptic FCN encodes each object instance or stuff category into a specific kernel weight with the proposed kernel generator and produces the prediction by convolving the high-resolution feature directly. That means it produces the results in a simple generate-kernel-then-segment workflow.

The main difference between Panoptic FCN and SOLO V2 lies in that it represents Things and Stuff in a unified manner (encode each instance into a specific kernel), while SOLO V2 represents Things and Stuff differently (Things by locations and a separate branch for Stuff segmentation).

CuberrChen commented 3 years ago

The answer is very clear, thank you！