CIDEr: Consensus-based Image Description Evaluation
Consensus Sentences (bold) captured by the proposed CIDEr evaluation protocol.
People:
Abstract:
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.
Paper
Ramakrishna Vedantam, C. Lawrence Zitnick, Devi Parikh
Consensus-based Image Description Evaluation
in proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015
[PDF] [Poster] [Extended Abstract]
Bibtex
@InProceedings{Vedantam_2015_CVPR,
author = {Vedantam, Ramakrishna and Lawrence Zitnick, C. and Parikh, Devi},
title = {CIDEr: Consensus-Based Image Description Evaluation},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2015}
}
Datasets
Download the PASCAL-50S, ABSTRACT-50S datasets here and the consensus annotations here
NOTE: '.mat' files can be easily loaded in Python using scipy.io.loadmat
Miscellanous
ArXiv version with supplementary material arXiv
Presentations
Berkeley Vision and Language Workshop Slides