Image Captioning with Context-Aware Auxiliary Guidance
This work offers an incremental improvement for researchers and practitioners working on image captioning models by enhancing their ability to incorporate global contextual information.
This paper addresses the limitation of encoder-decoder image captioning models that rely solely on past generated words by introducing a Context-Aware Auxiliary Guidance (CAAG) mechanism. CAAG helps the model perceive global contexts by using semantic attention on global predictions to refine current word generation, achieving competitive performance with 132.2 CIDEr-D on the Karpathy split and 130.7 CIDEr-D (c40) on the official COCO online evaluation server.
Image captioning is a challenging computer vision task, which aims to generate a natural language description of an image. Most recent researches follow the encoder-decoder framework which depends heavily on the previous generated words for the current prediction. Such methods can not effectively take advantage of the future predicted information to learn complete semantics. In this paper, we propose Context-Aware Auxiliary Guidance (CAAG) mechanism that can guide the captioning model to perceive global contexts. Upon the captioning model, CAAG performs semantic attention that selectively concentrates on useful information of the global predictions to reproduce the current generation. To validate the adaptability of the method, we apply CAAG to three popular captioners and our proposal achieves competitive performance on the challenging Microsoft COCO image captioning benchmark, e.g. 132.2 CIDEr-D score on Karpathy split and 130.7 CIDEr-D (c40) score on official online evaluation server.