Towards Non-I.I.D. image classification: A dataset and baselines

作者：

Highlights：

• We investigate the Non-I.I.D problem in image classification and give an index NI to measure the degree of distribution shift.

• We construct and release a Non-I.I.D. image dataset called NICO, which makes use of contexts to create various Non-IIDness exibly and consciously.

• We propose a novel model CNBB with batch balancing module as a baseline of exploiting CNN for general Non-I.I.D. image classification.

• Extensive experiments prove the capacity of NICO and the superiority of CNBB.

摘要

•We investigate the Non-I.I.D problem in image classification and give an index NI to measure the degree of distribution shift.•We construct and release a Non-I.I.D. image dataset called NICO, which makes use of contexts to create various Non-IIDness exibly and consciously.•We propose a novel model CNBB with batch balancing module as a baseline of exploiting CNN for general Non-I.I.D. image classification.•Extensive experiments prove the capacity of NICO and the superiority of CNBB.

论文关键词：Non-I.I.D,Dataset,Context,Bias,ConvNet,Batch balancing

论文评审过程：Received 14 July 2019, Revised 25 March 2020, Accepted 15 April 2020, Available online 2 June 2020, Version of Record 1 November 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107383