This projects detect Anomalous Behavior through live CCTV camera feed to alert the police or local authority for faster response time. We are using Spatio Temporal AutoEncoder and more importantly three models from Keras ie; Convolutional 3D, Convolutional 2D LSTM and Convolutional 3D Transpose. 👮‍♂️👮‍♀️📹🔍🔫⚖

Live Cell Histology: Extracting latent features from label-free live cell images using Adversarial Autoencoders

Towards Generative Modeling from (variational) Autoencoder to DCGAN

Filtering out the noise presented in the image by auto-enconder algorithm in TensorFow and Keras. Rare images, unclean crime images,medical noise images can be denoised and find out the desired outcome by using auto-encoders.

Autoencoder is a type of neural network where the output layer has the same dimensionality as the input layer. In simpler words, the number of output units in the output layer is equal to the number of input units in the input layer. An autoencoder replicates the data from the input to the output in an unsupervised manner and is therefore someti…

Colorizes grayscale images using a loaded model and displays original and predicted colorized versions.

In this research work, unsupervised abnormality has been detected by using intelligent and heterogeneous autonomous systems.

Live Cell Histology: Extracting latent features from label-free live cell images using Adversarial Autoencoders

vanilla and convolutional autoencoder for generating mnist images

Person Segmentation using custom Autoencoder architecture and evaluation using IoU and Dice metrics, will also include Unet architecture in the future.

Pytorch implementation of Masked Autoencoder I

Uma abordagem prática para construção de autoencoders convolucionais.