Accuracy - All Images: 50.5Accuracy - Corrupted Images: 45.8Accuracy - Clean Images: 53.2
Papers with Code · CC BY-SA 4.0
Domain Generalization
The idea of Domain Generalization is to learn from one or multiple training domains, to extract a domain-agnostic model which can be applied to an unseen domain
<span class="description-source">Source: Diagram Image Retrieval using Sketch-Based Deep Learning and Transfer Learning </span>
논문
Accuracy - All Images: 45.8Accuracy - Corrupted Images: 39.3Accuracy - Clean Images: 50.7
Accuracy - All Images: 47.5Accuracy - Corrupted Images: 43.3Accuracy - Clean Images: 51.3
Accuracy - All Images: 45Accuracy - Corrupted Images: 38.9Accuracy - Clean Images: 48.7
Accuracy - All Images: 53.5Accuracy - Corrupted Images: 46.9Accuracy - Clean Images: 56
Accuracy - All Images: 51.7Accuracy - Corrupted Images: 48.1Accuracy - Clean Images: 54.8
Accuracy - All Images: 49
Accuracy - All Images: 42.2Accuracy - Corrupted Images: 35.9Accuracy - Clean Images: 46.4
Accuracy - All Images: 42.6Accuracy - Corrupted Images: 34.9Accuracy - Clean Images: 47.5
Average Accuracy: 74.46
mIoU: 70.21
Accuracy: 82.8
Accuracy: 78.9
Accuracy - All Images: 42.8Accuracy - Corrupted Images: 37Accuracy - Clean Images: 47.3
Mean IoU: 31.2
Accuracy - All Images: 48.9Accuracy - Corrupted Images: 39.1Accuracy - Clean Images: 44.4
Accuracy - All Images: 57.2Accuracy - Corrupted Images: 51.8Accuracy - Clean Images: 59.7