huangeddie/ML-Aesthetics-NIMA
A PyTorch NIMA Implementation using DenseNet. NIMA is a research endeavor that rates the aesthetic quality of images
Jupyter Notebook
Issues
- 3
- 0
hyper-parameter
#19 opened by huanmofazong - 0
Results not align with the teaser
#18 opened by SmithLiu95 - 0
Neural Visualization
#14 opened by huangeddie - 0
- 0
improve readme
#16 opened by huangeddie - 0
Apply horizontal flipping on datasets
#3 opened by huangeddie - 0
- 0
Loss function: Earth mover's distance
#4 opened by huangeddie - 0
Train on a pretrained model (VGG16)
#11 opened by Xiangmingchen - 0
Resolve PIL warnings
#12 opened by huangeddie - 0
Simple binary classifier
#10 opened by huangeddie - 0
Binary classification
#6 opened by huangeddie - 0
Process bad images with Pillow
#9 opened by huangeddie - 0
save models
#8 opened by huangeddie - 0
More complex model
#7 opened by huangeddie - 0
model evaluation: linear correlation between mean of predicted pdf and mean of true pdf
#5 opened by huangeddie - 0
Train a vector quantization model
#2 opened by huangeddie - 0
Process meta features of images
#1 opened by huangeddie