1 paper
Simao Herdade, Armin Kappeler, Kofi Boakye +1
Image captioning models typically follow an encoder-decoder architecture which uses abstract image feature vectors as input to the encoder. One of the most successful algorithms us…