Google Machine Learning Models for Image Captioning Ported to TensorFlow and Open-Sourced

Google chronicled their journey over the past few years with their announcement around open-sourcing a TensorFlow model for image captioning, and some of the testing for comparing accuracy and performance benchmarks between the new approach and existing implementations. The 2014 Inception V1, 2015 Inception V2, and recently the Inception V3 model are improvements over one another, at 89.6, 91.8 and 93.9 percent in top-5 accuracy against an ImageNet 2012 image classification task. The BLEU-4 metric is used to measure quality of the machine generated captions by measuring the accuracy of sentence translation from one natural language to another. The TensorFlow-based approach took a 2 point gain over the previous leading model, DistBelief. One of the problems noted in porting and improving the new models from previous implementations is the process  of object classification in an image versus describing and relating the objects in an image…


Link to Full Article: Google Machine Learning Models for Image Captioning Ported to TensorFlow and Open-Sourced

Pin It on Pinterest

Share This

Join Our Newsletter

Sign up to our mailing list to receive the latest news and updates about homeAI.info and the Informed.AI Network of AI related websites which includes Events.AI, Neurons.AI, Awards.AI, and Vocation.AI

You have Successfully Subscribed!