(aside image)

Software

DeepCaption

Pytorch implementation of NeuralTalk with some improvments.
Code on GitHub

NeuralTalkTheano

Theano implementation of the NeuralTalk code by Karpathy et al.
Code on GitHub

PicSOM

Since 1998 we have been developing the PicSOM system - a comprehensive framework for content-based image and video retrieval research. The research has been and will be three-fold: algorithm development, international evaluations and applications.
Recent PicSOM releases can be found on GitHub.

SLMotion

SLMotion sign language oriented video analysis tool

ObjectDetection (od)

od-1.0.tar.gz

See example input data and detection result and heat map videos.