Image captioning problem: Given an image, describe what is happening in the image.
What is the state of the art work on image captioning?
Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It only takes a minute to sign up.Sign up to join this community
The state of the art works on image captioning problem can be found on 'Image Captioning Challege' with MSCOCO Dataset here.
Currently, 'Tencent', A Chinese giant is on the top with an approach based on Encoder-Decoder framework. It uses deep CNNs as encoders to encode the images into a vector that contains semantic information of the image and LSTMs to decode them. They also use multi-stage attention mechanism during decoding and Reinforcement Learning to directly optimize the non-differential functions and scores like CIDEr.
Update Oct 2020:
A lot have changed in the last 2.5 years, In addition to that, Now I'd recommend this source (papers with code) or this (madewithml) for literature review of SOTA papers as well as their github implementations.