5,337
edits
Visual Learning and Recognition: Difference between revisions
Visual Learning and Recognition (view source)
Revision as of 16:43, 22 October 2020
, 22 October 2020→Summary of online hard example mining
| Line 677: | Line 677: | ||
#* Filter duplicates using NMS. | #* Filter duplicates using NMS. | ||
# Backwards pass only on the top <math>B/N</math> examples. | # Backwards pass only on the top <math>B/N</math> examples. | ||
===Bells & Whistles=== | |||
Most used: | |||
* Multi-scale training and testing | |||
* Iterative bounding-box prediction | |||
* Ensemble | |||
* More data | |||
Exploratory | |||
* Loss variants | |||
* Network variants | |||
* NMS variants | |||
==Will be on the exam== | ==Will be on the exam== | ||