Exploring the discriminating power of different loss functions for classification

2D/3D t-SNE Model Analysis

Use your classification neural network for object detection and localization

code to help with tsne plotting

An image search implementation in python using tensorflow keras, scikit-learn, scipy and matplotlib.

3D T-SNE graphs with sliders and checkboxes to visualize the T-SNE cloud at every epoch for specific labels. Optionally you can also track specific datapoint by labeling it with a unique marker.

Deepfakes Video classification via CNN, LSTM, C3D and triplets [IWBF'20]