5,337
edits
Visual Learning and Recognition: Difference between revisions
Visual Learning and Recognition (view source)
Revision as of 17:02, 20 October 2020
, 20 October 2020→Mask R-CNN
| Line 615: | Line 615: | ||
===Mask R-CNN=== | ===Mask R-CNN=== | ||
Given an image, | * Given an image, calculate fully-conv features. | ||
* Extract feature map for a region using RoIAlign. | |||
* Do fully-conv over the region only to get segmentation + class. | |||
===RoI Pooling variants=== | |||
* RoI Pool (SPPNet, Fast/Faster R-CNN) | |||
* RoI Warp (MNC) | |||
* RoI Align (Mask R-CNN) | |||
==Will be on the exam== | ==Will be on the exam== | ||