CIDEr: Consensus-based Image Description Evaluation

Ramakrishna Vedantam,  C. Lawrence Zitnick,  Devi Parikh

(Submitted on 20 Nov 2014 ( v1), last revised 3 Jun 2015 (this version, v2))

Automatically describing an image with a sentence is a long-standing challenge in computer vision and natural language processing. Due to recent progress in object detection, attribute classification, action recognition, etc., there is renewed interest in this area. However, evaluating the quality of descriptions has proven to be challenging. We propose a novel paradigm for evaluating image descriptions that uses human consensus. This paradigm consists of three main parts: a new triplet-based method of collecting human annotations to measure consensus, a new automated metric (CIDEr) that captures consensus, and two new datasets: PASCAL-50S and ABSTRACT-50S that contain 50 sentences describing each image. Our simple metric captures human judgment of consensus better than existing metrics across sentences generated by various sources. We also evaluate five state-of-the-art image description approaches using this new protocol and provide a benchmark for future comparisons. A version of CIDEr named CIDEr-D is available as a part of MS COCO evaluation server to enable systematic evaluation and benchmarking.

Comments:	To appear in CVPR 2015
Subjects:	Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Information Retrieval (cs.IR)
Cite as:	arXiv:1411.5726 [cs.CV]
	(or arXiv:1411.5726v2 [cs.CV] for this version)

Submission history

From: Ramakrishna Vedantam [ view email] 
[v1] Thu, 20 Nov 2014 23:54:35 GMT (1980kb,D)
[v2] Wed, 3 Jun 2015 01:42:20 GMT (12171kb,D)