File size: 1,570 Bytes
6b9c6e5 0db6482 d999b19 5521118 0db6482 255035e 0db6482 255035e 0db6482 f92dbd7 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 |
---
license: apache-2.0
---
# Classifier architecture
The classifier uses DenseNet161 as the encoder and some linear layers at classifier base.
# Model accuracy:
Model achieves 91.3% accuracy on the validation set. \
F1-score per class: {'digital': 0.9873773235685747, 'hard': 0.9338602782753218, 'soft': 0.8444277483052108} \
Mean F1-score: 0.9218884500497024 \
Accuracy: 0.913
# Training dataset metadata:
1. Dataset classes: ['soft', 'digital', 'hard']
2. Number of classes: 3
3. Total number of images: 18415
# Number of images per class:
- soft : 5482
- digital : 1206
- hard : 11727
# Classes description:
1. The **hard** class denotes a group of scenes to which a coarser background removal method should be applied, intended for objects with an edge without small details.
The hard class contains the following categories of objects:
object, laptop, charger, pc mouse, pc, rocks, table, bed, box, sneakers, ship, wire, guitar, fork, spoon, plate, keyboard, car, bus, screwdriver, ball, door, flower, clocks, fruit , food, robot.
2. The **soft** class denotes a group of scenes to which you want to apply a soft background removal method intended for people, hair, clothes, and other similar types of objects. The soft class contains the following categories of objects:
animal, people, human, man, woman, t-shirt, hairs, hair, dog, cat, monkey, cow, medusa, clothes
3. The **digital** class denotes a group of images with digital graphics, such as screenshots, logos, and so on.
The digital class contains the following categories of scenes:
screenshot
|